The R Project SVN R

Rev

Rev 28322 | Only display areas with differences | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed

Rev 28322 Rev 28348
1 
<html lang="en">
 1 
<html lang="en">
2 
<head>
 2 
<head>
3 
<title>R FAQ</title>
 3 
<title>R FAQ</title>
4 
<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
 4 
<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
5 
<meta name="description" content="R FAQ">
 5 
<meta name="description" content="R FAQ">
6 
<meta name="generator" content="makeinfo 4.6">
 6 
<meta name="generator" content="makeinfo 4.6">
7 
<meta http-equiv="Content-Style-Type" content="text/css">
 7 
<meta http-equiv="Content-Style-Type" content="text/css">
8 
<style type="text/css"><!--
 8 
<style type="text/css"><!--
9 
  pre.display { font-family:inherit }
 9 
  pre.display { font-family:inherit }
10 
  pre.format  { font-family:inherit }
 10 
  pre.format  { font-family:inherit }
11 
  pre.smalldisplay { font-family:inherit; font-size:smaller }
 11 
  pre.smalldisplay { font-family:inherit; font-size:smaller }
12 
  pre.smallformat  { font-family:inherit; font-size:smaller }
 12 
  pre.smallformat  { font-family:inherit; font-size:smaller }
13 
  pre.smallexample { font-size:smaller }
 13 
  pre.smallexample { font-size:smaller }
14 
  pre.smalllisp    { font-size:smaller }
 14 
  pre.smalllisp    { font-size:smaller }
15 
--></style>
 15 
--></style>
16 
</head>
 16 
</head>
17 
<body>
 17 
<body>
18 
<h1 class="settitle">R FAQ</h1>
 18 
<h1 class="settitle">R FAQ</h1>
19 
<div class="node">
 19 
<div class="node">
20 
<p><hr>
 20 
<p><hr>
21 
Node:&nbsp;<a name="Top">Top</a>,
 21 
Node:&nbsp;<a name="Top">Top</a>,
22 
Next:&nbsp;<a rel="next" accesskey="n" href="#Introduction">Introduction</a>,
 22 
Next:&nbsp;<a rel="next" accesskey="n" href="#Introduction">Introduction</a>,
23 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#dir">(dir)</a>,
 23 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#dir">(dir)</a>,
24 
Up:&nbsp;<a rel="up" accesskey="u" href="#dir">(dir)</a>
 24 
Up:&nbsp;<a rel="up" accesskey="u" href="#dir">(dir)</a>
25 
<br>
 25 
<br>
26 
</div>
 26 
</div>
27 
 
 27 
 
28 
<h2 class="unnumbered">R FAQ</h2>
 28 
<h2 class="unnumbered">R FAQ</h2>
29 
 
 29 
 
30 
<h2>Frequently Asked Questions on R</h2>
 30 
<h2>Frequently Asked Questions on R</h2>
31 
<h2>Version 1.8-41, 2004-02-16</h2>
 31 
<h2>Version 1.8-42, 2004-02-19</h2>
32 
<h2>ISBN 3-900051-01-1</h2>
 32 
<h2>ISBN 3-900051-01-1</h2>
33 
<address>Kurt Hornik</address>
 33 
<address>Kurt Hornik</address>
34 
<p><p><hr><p>
 34 
<p><p><hr><p>
35 
 
 35 
 
36 
<ul class="menu">
 36 
<ul class="menu">
37 
<li><a accesskey="1" href="#Introduction">Introduction</a>:
37 
<li><a accesskey="1" href="#Introduction">Introduction</a>:
38 
<li><a accesskey="2" href="#R%20Basics">R Basics</a>:
38 
<li><a accesskey="2" href="#R%20Basics">R Basics</a>:
39 
<li><a accesskey="3" href="#R%20and%20S">R and S</a>:
39 
<li><a accesskey="3" href="#R%20and%20S">R and S</a>:
40 
<li><a accesskey="4" href="#R%20Web%20Interfaces">R Web Interfaces</a>:
40 
<li><a accesskey="4" href="#R%20Web%20Interfaces">R Web Interfaces</a>:
41 
<li><a accesskey="5" href="#R%20Add-On%20Packages">R Add-On Packages</a>:
41 
<li><a accesskey="5" href="#R%20Add-On%20Packages">R Add-On Packages</a>:
42 
<li><a accesskey="6" href="#R%20and%20Emacs">R and Emacs</a>:
42 
<li><a accesskey="6" href="#R%20and%20Emacs">R and Emacs</a>:
43 
<li><a accesskey="7" href="#R%20Miscellanea">R Miscellanea</a>:
43 
<li><a accesskey="7" href="#R%20Miscellanea">R Miscellanea</a>:
44 
<li><a accesskey="8" href="#R%20Programming">R Programming</a>:
44 
<li><a accesskey="8" href="#R%20Programming">R Programming</a>:
45 
<li><a accesskey="9" href="#R%20Bugs">R Bugs</a>:
45 
<li><a accesskey="9" href="#R%20Bugs">R Bugs</a>:
46 
<li><a href="#Acknowledgments">Acknowledgments</a>:
46 
<li><a href="#Acknowledgments">Acknowledgments</a>:
47 
</ul>
 47 
</ul>
48 
 
 48 
 
49 
<div class="node">
 49 
<div class="node">
50 
<p><hr>
 50 
<p><hr>
51 
Node:&nbsp;<a name="Introduction">Introduction</a>,
 51 
Node:&nbsp;<a name="Introduction">Introduction</a>,
52 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Basics">R Basics</a>,
 52 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Basics">R Basics</a>,
53 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Top">Top</a>,
 53 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Top">Top</a>,
54 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 54 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
55 
<br>
 55 
<br>
56 
</div>
 56 
</div>
57 
 
 57 
 
58 
<h2 class="chapter">1 Introduction</h2>
 58 
<h2 class="chapter">1 Introduction</h2>
59 
 
 59 
 
60 
<p>This document contains answers to some of the most frequently asked
 60 
<p>This document contains answers to some of the most frequently asked
61 
questions about R.
 61 
questions about R.
62 
 
 62 
 
63 
<ul class="menu">
 63 
<ul class="menu">
64 
<li><a accesskey="1" href="#Legalese">Legalese</a>:
64 
<li><a accesskey="1" href="#Legalese">Legalese</a>:
65 
<li><a accesskey="2" href="#Obtaining%20this%20document">Obtaining this document</a>:
65 
<li><a accesskey="2" href="#Obtaining%20this%20document">Obtaining this document</a>:
66 
<li><a accesskey="3" href="#Citing%20this%20document">Citing this document</a>:
66 
<li><a accesskey="3" href="#Citing%20this%20document">Citing this document</a>:
67 
<li><a accesskey="4" href="#Notation">Notation</a>:
67 
<li><a accesskey="4" href="#Notation">Notation</a>:
68 
<li><a accesskey="5" href="#Feedback">Feedback</a>:
68 
<li><a accesskey="5" href="#Feedback">Feedback</a>:
69 
</ul>
 69 
</ul>
70 
 
 70 
 
71 
<div class="node">
 71 
<div class="node">
72 
<p><hr>
 72 
<p><hr>
73 
Node:&nbsp;<a name="Legalese">Legalese</a>,
 73 
Node:&nbsp;<a name="Legalese">Legalese</a>,
74 
Next:&nbsp;<a rel="next" accesskey="n" href="#Obtaining%20this%20document">Obtaining this document</a>,
 74 
Next:&nbsp;<a rel="next" accesskey="n" href="#Obtaining%20this%20document">Obtaining this document</a>,
75 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Introduction">Introduction</a>,
 75 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Introduction">Introduction</a>,
76 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
 76 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
77 
<br>
 77 
<br>
78 
</div>
 78 
</div>
79 
 
 79 
 
80 
<h3 class="section">1.1 Legalese</h3>
 80 
<h3 class="section">1.1 Legalese</h3>
81 
 
 81 
 
82 
<p>This document is copyright &copy; 1998-2004 by Kurt
 82 
<p>This document is copyright &copy; 1998-2004 by Kurt
83 
Hornik.
 83 
Hornik.
84 
 
 84 
 
85 
   <p>This document is free software; you can redistribute it and/or modify it
 85 
   <p>This document is free software; you can redistribute it and/or modify it
86 
under the terms of the <small>GNU</small> General Public License as published
 86 
under the terms of the <small>GNU</small> General Public License as published
87 
by the Free Software Foundation; either version 2, or (at your option)
 87 
by the Free Software Foundation; either version 2, or (at your option)
88 
any later version.
 88 
any later version.
89 
 
 89 
 
90 
   <p>This document is distributed in the hope that it will be useful, but
 90 
   <p>This document is distributed in the hope that it will be useful, but
91 
WITHOUT ANY WARRANTY; without even the implied warranty of
 91 
WITHOUT ANY WARRANTY; without even the implied warranty of
92 
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 92 
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
93 
<small>GNU</small> General Public License for more details.
 93 
<small>GNU</small> General Public License for more details.
94 
 
 94 
 
95 
   <p>A copy of the <small>GNU</small> General Public License is available via WWW
 95 
   <p>A copy of the <small>GNU</small> General Public License is available via WWW
96 
at
 96 
at
97 
 
 97 
 
98 
<pre class="display">     <a href="http://www.gnu.org/copyleft/gpl.html">http://www.gnu.org/copyleft/gpl.html</a>.
 98 
<pre class="display">     <a href="http://www.gnu.org/copyleft/gpl.html">http://www.gnu.org/copyleft/gpl.html</a>.
99 
     </pre>
 99 
     </pre>
100 
 
 100 
 
101 
<p>You can also obtain it by writing to the Free Software Foundation, Inc.,
 101 
<p>You can also obtain it by writing to the Free Software Foundation, Inc.,
102 
59 Temple Place -- Suite 330, Boston, MA 02111-1307, USA.
 102 
59 Temple Place -- Suite 330, Boston, MA 02111-1307, USA.
103 
 
 103 
 
104 
<div class="node">
 104 
<div class="node">
105 
<p><hr>
 105 
<p><hr>
106 
Node:&nbsp;<a name="Obtaining%20this%20document">Obtaining this document</a>,
 106 
Node:&nbsp;<a name="Obtaining%20this%20document">Obtaining this document</a>,
107 
Next:&nbsp;<a rel="next" accesskey="n" href="#Citing%20this%20document">Citing this document</a>,
 107 
Next:&nbsp;<a rel="next" accesskey="n" href="#Citing%20this%20document">Citing this document</a>,
108 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Legalese">Legalese</a>,
 108 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Legalese">Legalese</a>,
109 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
 109 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
110 
<br>
 110 
<br>
111 
</div>
 111 
</div>
112 
 
 112 
 
113 
<h3 class="section">1.2 Obtaining this document</h3>
 113 
<h3 class="section">1.2 Obtaining this document</h3>
114 
 
 114 
 
115 
<p>The latest version of this document is always available from
 115 
<p>The latest version of this document is always available from
116 
 
 116 
 
117 
<pre class="display">     <a href="http://www.ci.tuwien.ac.at/~hornik/R/">http://www.ci.tuwien.ac.at/~hornik/R/</a>
 117 
<pre class="display">     <a href="http://www.ci.tuwien.ac.at/~hornik/R/">http://www.ci.tuwien.ac.at/~hornik/R/</a>
118 
     </pre>
 118 
     </pre>
119 
 
 119 
 
120 
   <p>From there, you can obtain versions converted to
 120 
   <p>From there, you can obtain versions converted to
121 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.txt">plain <small>ASCII</small> text</a>,
 121 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.txt">plain <small>ASCII</small> text</a>,
122 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.dvi.gz">DVI</a>,
 122 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.dvi.gz">DVI</a>,
123 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.info.gz"><small>GNU</small> info</a>, <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html"><small>HTML</small></a>,
 123 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.info.gz"><small>GNU</small> info</a>, <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html"><small>HTML</small></a>,
124 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.pdf">PDF</a>,
 124 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.pdf">PDF</a>,
125 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.ps.gz">PostScript</a> as
 125 
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.ps.gz">PostScript</a> as
126 
well as the <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.texi">Texinfo source</a> used for creating all these formats using the
 126 
well as the <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.texi">Texinfo source</a> used for creating all these formats using the
127 
<a href="http://texinfo.org/"><small>GNU</small> Texinfo system</a>.
 127 
<a href="http://texinfo.org/"><small>GNU</small> Texinfo system</a>.
128 
 
 128 
 
129 
   <p>You can also obtain the R <small>FAQ</small> from the <code>doc/FAQ</code>
 129 
   <p>You can also obtain the R <small>FAQ</small> from the <code>doc/FAQ</code>
130 
subdirectory of a <small>CRAN</small> site (see <a href="#What%20is%20CRAN%3f">What is CRAN?</a>).
 130 
subdirectory of a <small>CRAN</small> site (see <a href="#What%20is%20CRAN%3f">What is CRAN?</a>).
131 
 
 131 
 
132 
<div class="node">
 132 
<div class="node">
133 
<p><hr>
 133 
<p><hr>
134 
Node:&nbsp;<a name="Citing%20this%20document">Citing this document</a>,
 134 
Node:&nbsp;<a name="Citing%20this%20document">Citing this document</a>,
135 
Next:&nbsp;<a rel="next" accesskey="n" href="#Notation">Notation</a>,
 135 
Next:&nbsp;<a rel="next" accesskey="n" href="#Notation">Notation</a>,
136 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Obtaining%20this%20document">Obtaining this document</a>,
 136 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Obtaining%20this%20document">Obtaining this document</a>,
137 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
 137 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
138 
<br>
 138 
<br>
139 
</div>
 139 
</div>
140 
 
 140 
 
141 
<h3 class="section">1.3 Citing this document</h3>
 141 
<h3 class="section">1.3 Citing this document</h3>
142 
 
 142 
 
143 
<p>In publications, please refer to this <small>FAQ</small> as Hornik
 143 
<p>In publications, please refer to this <small>FAQ</small> as Hornik
144 
(2004), "The R <small>FAQ</small>", and give the above,
 144 
(2004), "The R <small>FAQ</small>", and give the above,
145 
<em>official</em> <small>URL</small> and the ISBN 3-900051-01-1.
 145 
<em>official</em> <small>URL</small> and the ISBN 3-900051-01-1.
146 
 
 146 
 
147 
<div class="node">
 147 
<div class="node">
148 
<p><hr>
 148 
<p><hr>
149 
Node:&nbsp;<a name="Notation">Notation</a>,
 149 
Node:&nbsp;<a name="Notation">Notation</a>,
150 
Next:&nbsp;<a rel="next" accesskey="n" href="#Feedback">Feedback</a>,
 150 
Next:&nbsp;<a rel="next" accesskey="n" href="#Feedback">Feedback</a>,
151 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Citing%20this%20document">Citing this document</a>,
 151 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Citing%20this%20document">Citing this document</a>,
152 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
 152 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
153 
<br>
 153 
<br>
154 
</div>
 154 
</div>
155 
 
 155 
 
156 
<h3 class="section">1.4 Notation</h3>
 156 
<h3 class="section">1.4 Notation</h3>
157 
 
 157 
 
158 
<p>Everything should be pretty standard.  <code>R&gt;</code> is used for the R
 158 
<p>Everything should be pretty standard.  <code>R&gt;</code> is used for the R
159 
prompt, and a <code>$</code> for the shell prompt (where applicable).
 159 
prompt, and a <code>$</code> for the shell prompt (where applicable).
160 
 
 160 
 
161 
<div class="node">
 161 
<div class="node">
162 
<p><hr>
 162 
<p><hr>
163 
Node:&nbsp;<a name="Feedback">Feedback</a>,
 163 
Node:&nbsp;<a name="Feedback">Feedback</a>,
164 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Notation">Notation</a>,
 164 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Notation">Notation</a>,
165 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
 165 
Up:&nbsp;<a rel="up" accesskey="u" href="#Introduction">Introduction</a>
166 
<br>
 166 
<br>
167 
</div>
 167 
</div>
168 
 
 168 
 
169 
<h3 class="section">1.5 Feedback</h3>
 169 
<h3 class="section">1.5 Feedback</h3>
170 
 
 170 
 
171 
<p>Feedback is of course most welcome.
 171 
<p>Feedback is of course most welcome.
172 
 
 172 
 
173 
   <p>In particular, note that I do not have access to Windows or Macintosh
 173 
   <p>In particular, note that I do not have access to Windows or Macintosh
174 
systems.  Features specific to the Windows and MacOS X ports of R are
 174 
systems.  Features specific to the Windows and MacOS X ports of R are
175 
described in the <a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">"R for Windows <small>FAQ</small>"</a> and the
 175 
described in the <a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">"R for Windows <small>FAQ</small>"</a> and the
176 
<a href="http://cran.r-project.org/bin/macosx/RAqua-FAQ.html">"R for Macintosh <small>FAQ</small>/DOC"</a>.  If you have information on Macintosh or
 176 
<a href="http://cran.r-project.org/bin/macosx/RAqua-FAQ.html">"R for Macintosh <small>FAQ</small>/DOC"</a>.  If you have information on Macintosh or
177 
Windows systems that you think should be added to this document, please
 177 
Windows systems that you think should be added to this document, please
178 
let me know.
 178 
let me know.
179 
 
 179 
 
180 
<div class="node">
 180 
<div class="node">
181 
<p><hr>
 181 
<p><hr>
182 
Node:&nbsp;<a name="R%20Basics">R Basics</a>,
 182 
Node:&nbsp;<a name="R%20Basics">R Basics</a>,
183 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20and%20S">R and S</a>,
 183 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20and%20S">R and S</a>,
184 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Introduction">Introduction</a>,
 184 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Introduction">Introduction</a>,
185 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 185 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
186 
<br>
 186 
<br>
187 
</div>
 187 
</div>
188 
 
 188 
 
189 
<h2 class="chapter">2 R Basics</h2>
 189 
<h2 class="chapter">2 R Basics</h2>
190 
 
 190 
 
191 
<ul class="menu">
 191 
<ul class="menu">
192 
<li><a accesskey="1" href="#What%20is%20R%3f">What is R?</a>:
192 
<li><a accesskey="1" href="#What%20is%20R%3f">What is R?</a>:
193 
<li><a accesskey="2" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>:
193 
<li><a accesskey="2" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>:
194 
<li><a accesskey="3" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>:
194 
<li><a accesskey="3" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>:
195 
<li><a accesskey="4" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>:
195 
<li><a accesskey="4" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>:
196 
<li><a accesskey="5" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>:
196 
<li><a accesskey="5" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>:
197 
<li><a accesskey="6" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>:
197 
<li><a accesskey="6" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>:
198 
<li><a accesskey="7" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>:
198 
<li><a accesskey="7" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>:
199 
<li><a accesskey="8" href="#Citing%20R">Citing R</a>:
199 
<li><a accesskey="8" href="#Citing%20R">Citing R</a>:
200 
<li><a accesskey="9" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>:
200 
<li><a accesskey="9" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>:
201 
<li><a href="#What%20is%20CRAN%3f">What is CRAN?</a>:
201 
<li><a href="#What%20is%20CRAN%3f">What is CRAN?</a>:
202 
<li><a href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>:
202 
<li><a href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>:
203 
<li><a href="#Why%20is%20R%20named%20R%3f">Why is R named R?</a>:
203 
<li><a href="#Why%20is%20R%20named%20R%3f">Why is R named R?</a>:
204 
</ul>
 204 
</ul>
205 
 
 205 
 
206 
<div class="node">
 206 
<div class="node">
207 
<p><hr>
 207 
<p><hr>
208 
Node:&nbsp;<a name="What%20is%20R%3f">What is R?</a>,
 208 
Node:&nbsp;<a name="What%20is%20R%3f">What is R?</a>,
209 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
 209 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
210 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Basics">R Basics</a>,
 210 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Basics">R Basics</a>,
211 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 211 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
212 
<br>
 212 
<br>
213 
</div>
 213 
</div>
214 
 
 214 
 
215 
<h3 class="section">2.1 What is R?</h3>
 215 
<h3 class="section">2.1 What is R?</h3>
216 
 
 216 
 
217 
<p>R is a system for statistical computation and graphics.  It consists of
 217 
<p>R is a system for statistical computation and graphics.  It consists of
218 
a language plus a run-time environment with graphics, a debugger, access
 218 
a language plus a run-time environment with graphics, a debugger, access
219 
to certain system functions, and the ability to run programs stored in
 219 
to certain system functions, and the ability to run programs stored in
220 
script files.
 220 
script files.
221 
 
 221 
 
222 
   <p>The design of R has been heavily influenced by two existing languages:
 222 
   <p>The design of R has been heavily influenced by two existing languages:
223 
Becker, Chambers &amp; Wilks' S (see <a href="#What%20is%20S%3f">What is S?</a>) and Sussman's
 223 
Becker, Chambers &amp; Wilks' S (see <a href="#What%20is%20S%3f">What is S?</a>) and Sussman's
224 
<a href="http://www.cs.indiana.edu/scheme-repository/home.html">Scheme</a>.
224 
<a href="http://www.cs.indiana.edu/scheme-repository/home.html">Scheme</a>.
225 
Whereas the resulting language is very similar in appearance to S, the
 225 
Whereas the resulting language is very similar in appearance to S, the
226 
underlying implementation and semantics are derived from Scheme.
226 
underlying implementation and semantics are derived from Scheme.
227 
See <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>, for further details.
 227 
See <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>, for further details.
228 
 
 228 
 
229 
   <p>The core of R is an interpreted computer language which allows branching
 229 
   <p>The core of R is an interpreted computer language which allows branching
230 
and looping as well as modular programming using functions.  Most of the
 230 
and looping as well as modular programming using functions.  Most of the
231 
user-visible functions in R are written in R.  It is possible for the
 231 
user-visible functions in R are written in R.  It is possible for the
232 
user to interface to procedures written in the C, C++, or FORTRAN
 232 
user to interface to procedures written in the C, C++, or FORTRAN
233 
languages for efficiency.  The R distribution contains functionality for
 233 
languages for efficiency.  The R distribution contains functionality for
234 
a large number of statistical procedures.  Among these are: linear and
 234 
a large number of statistical procedures.  Among these are: linear and
235 
generalized linear models, nonlinear regression models, time series
 235 
generalized linear models, nonlinear regression models, time series
236 
analysis, classical parametric and nonparametric tests, clustering and
 236 
analysis, classical parametric and nonparametric tests, clustering and
237 
smoothing.  There is also a large set of functions which provide a
 237 
smoothing.  There is also a large set of functions which provide a
238 
flexible graphical environment for creating various kinds of data
 238 
flexible graphical environment for creating various kinds of data
239 
presentations.  Additional modules ("add-on packages") are available
 239 
presentations.  Additional modules ("add-on packages") are available
240 
for a variety of specific purposes (see <a href="#R%20Add-On%20Packages">R Add-On Packages</a>).
 240 
for a variety of specific purposes (see <a href="#R%20Add-On%20Packages">R Add-On Packages</a>).
241 
 
 241 
 
242 
   <p>R was initially written by <a href="mailto:Ross.Ihaka@R-project.org">Ross Ihaka</a>
 242 
   <p>R was initially written by <a href="mailto:Ross.Ihaka@R-project.org">Ross Ihaka</a>
243 
and <a href="mailto:Robert.Gentleman@R-project.org">Robert Gentleman</a> at the
 243 
and <a href="mailto:Robert.Gentleman@R-project.org">Robert Gentleman</a> at the
244 
Department of Statistics of the University of Auckland in Auckland, New
 244 
Department of Statistics of the University of Auckland in Auckland, New
245 
Zealand.  In addition, a large group of individuals has contributed to R
 245 
Zealand.  In addition, a large group of individuals has contributed to R
246 
by sending code and bug reports.
 246 
by sending code and bug reports.
247 
 
 247 
 
248 
   <p>Since mid-1997 there has been a core group (the "R Core Team") who can
 248 
   <p>Since mid-1997 there has been a core group (the "R Core Team") who can
249 
modify the R source code CVS archive.  The group currently consists of
 249 
modify the R source code CVS archive.  The group currently consists of
250 
Doug Bates, John Chambers, Peter Dalgaard, Robert Gentleman, Kurt
 250 
Doug Bates, John Chambers, Peter Dalgaard, Robert Gentleman, Kurt
251 
Hornik, Stefano Iacus, Ross Ihaka, Friedrich Leisch, Thomas Lumley,
 251 
Hornik, Stefano Iacus, Ross Ihaka, Friedrich Leisch, Thomas Lumley,
252 
Martin Maechler, Duncan Murdoch, Paul Murrell, Martyn Plummer, Brian
 252 
Martin Maechler, Duncan Murdoch, Paul Murrell, Martyn Plummer, Brian
253 
Ripley, Duncan Temple Lang, and Luke Tierney.
 253 
Ripley, Duncan Temple Lang, and Luke Tierney.
254 
 
 254 
 
255 
   <p>R has a home page at <a href="http://www.R-project.org/">http://www.R-project.org/</a>.  It is free
 255 
   <p>R has a home page at <a href="http://www.R-project.org/">http://www.R-project.org/</a>.  It is free
256 
software distributed under a <small>GNU</small>-style copyleft, and an
 256 
software distributed under a <small>GNU</small>-style copyleft, and an
257 
official part of the <small>GNU</small> project ("<small>GNU</small> S").
 257 
official part of the <small>GNU</small> project ("<small>GNU</small> S").
258 
 
 258 
 
259 
<div class="node">
 259 
<div class="node">
260 
<p><hr>
 260 
<p><hr>
261 
Node:&nbsp;<a name="What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
 261 
Node:&nbsp;<a name="What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
262 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
 262 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
263 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20R%3f">What is R?</a>,
 263 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20R%3f">What is R?</a>,
264 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 264 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
265 
<br>
 265 
<br>
266 
</div>
 266 
</div>
267 
 
 267 
 
268 
<h3 class="section">2.2 What machines does R run on?</h3>
 268 
<h3 class="section">2.2 What machines does R run on?</h3>
269 
 
 269 
 
270 
<p>R is being developed for the Unix, Windows and Mac families of operating
 270 
<p>R is being developed for the Unix, Windows and Mac families of operating
271 
systems.  Support for Mac OS Classic will end with the 1.7 series.
 271 
systems.  Support for Mac OS Classic will end with the 1.7 series.
272 
 
 272 
 
273 
   <p>The current version of R will configure and build under a number of
 273 
   <p>The current version of R will configure and build under a number of
274 
common Unix platforms including i386-freebsd, <var>cpu</var>-linux-gnu for
 274 
common Unix platforms including i386-freebsd, <var>cpu</var>-linux-gnu for
275 
the i386, alpha, arm, hppa, ia64, m68k, powerpc, and sparc CPUs (see
 275 
the i386, alpha, arm, hppa, ia64, m68k, powerpc, and sparc CPUs (see
276 
e.g. <a href="http://buildd.debian.org/build.php?&amp;pkg=r-base">http://buildd.debian.org/build.php?&amp;pkg=r-base</a>),
 276 
e.g. <a href="http://buildd.debian.org/build.php?&amp;pkg=r-base">http://buildd.debian.org/build.php?&amp;pkg=r-base</a>),
277 
i386-sun-solaris, powerpc-apple-darwin, mips-sgi-irix, alpha-dec-osf4,
 277 
i386-sun-solaris, powerpc-apple-darwin, mips-sgi-irix, alpha-dec-osf4,
278 
rs6000-ibm-aix, hppa-hp-hpux, and sparc-sun-solaris.
 278 
rs6000-ibm-aix, hppa-hp-hpux, and sparc-sun-solaris.
279 
 
 279 
 
280 
   <p>If you know about other platforms, please drop us a note.
 280 
   <p>If you know about other platforms, please drop us a note.
281 
 
 281 
 
282 
<div class="node">
 282 
<div class="node">
283 
<p><hr>
 283 
<p><hr>
284 
Node:&nbsp;<a name="What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
 284 
Node:&nbsp;<a name="What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
285 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
 285 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
286 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
 286 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
287 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 287 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
288 
<br>
 288 
<br>
289 
</div>
 289 
</div>
290 
 
 290 
 
291 
<h3 class="section">2.3 What is the current version of R?</h3>
 291 
<h3 class="section">2.3 What is the current version of R?</h3>
292 
 
 292 
 
293 
<p>The current released version is 1.8.1.  Based on this
 293 
<p>The current released version is 1.8.1.  Based on this
294 
`major.minor.patchlevel' numbering scheme, there are two development
 294 
`major.minor.patchlevel' numbering scheme, there are two development
295 
versions of R, working towards the next patch (`r-patched') and minor or
 295 
versions of R, working towards the next patch (`r-patched') and minor or
296 
eventually major (`r-devel') releases of R, respectively.  Version
 296 
eventually major (`r-devel') releases of R, respectively.  Version
297 
r-patched is for bug fixes mostly.  New features are typically
 297 
r-patched is for bug fixes mostly.  New features are typically
298 
introduced in r-devel.
 298 
introduced in r-devel.
299 
 
 299 
 
300 
<div class="node">
 300 
<div class="node">
301 
<p><hr>
 301 
<p><hr>
302 
Node:&nbsp;<a name="How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
 302 
Node:&nbsp;<a name="How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
303 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
 303 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
304 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
 304 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
305 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 305 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
306 
<br>
 306 
<br>
307 
</div>
 307 
</div>
308 
 
 308 
 
309 
<h3 class="section">2.4 How can R be obtained?</h3>
 309 
<h3 class="section">2.4 How can R be obtained?</h3>
310 
 
 310 
 
311 
<p>Sources, binaries and documentation for R can be obtained via <small>CRAN</small>,
 311 
<p>Sources, binaries and documentation for R can be obtained via <small>CRAN</small>,
312 
the "Comprehensive R Archive Network" (see <a href="#What%20is%20CRAN%3f">What is CRAN?</a>).
 312 
the "Comprehensive R Archive Network" (see <a href="#What%20is%20CRAN%3f">What is CRAN?</a>).
313 
 
 313 
 
314 
   <p>Sources are also available via anonymous rsync.  Use
 314 
   <p>Sources are also available via anonymous rsync.  Use
315 
 
 315 
 
316 
<pre class="example">     rsync -rC --delete rsync.R-project.org::<var>module</var> R
 316 
<pre class="example">     rsync -rC --delete rsync.R-project.org::<var>module</var> R
317 
     </pre>
 317 
     </pre>
318 
 
 318 
 
319 
<p>to create a copy of the source tree specified by <var>module</var> in the
 319 
<p>to create a copy of the source tree specified by <var>module</var> in the
320 
subdirectory <code>R</code> of the current directory, where <var>module</var>
 320 
subdirectory <code>R</code> of the current directory, where <var>module</var>
321 
specifies one of the three existing flavors of the R sources, and can be
 321 
specifies one of the three existing flavors of the R sources, and can be
322 
one of <code>r-release</code> (current released version), <code>r-patched</code>
 322 
one of <code>r-release</code> (current released version), <code>r-patched</code>
323 
(patched released version), and <code>r-devel</code> (development version).
323 
(patched released version), and <code>r-devel</code> (development version).
324 
The rsync trees are created directly from the master CVS archive and are
 324 
The rsync trees are created directly from the master CVS archive and are
325 
updated hourly.  The <code>-C</code> and in the <code>rsync</code> command
 325 
updated hourly.  The <code>-C</code> and in the <code>rsync</code> command
326 
is to cause it to skip the CVS directories.  Further information on
 326 
is to cause it to skip the CVS directories.  Further information on
327 
<code>rsync</code> is available at <a href="http://rsync.samba.org/rsync/">http://rsync.samba.org/rsync/</a>.
 327 
<code>rsync</code> is available at <a href="http://rsync.samba.org/rsync/">http://rsync.samba.org/rsync/</a>.
328 
 
 328 
 
329 
   <p>The sources of the development version are also available via anonymous
 329 
   <p>The sources of the development version are also available via anonymous
330 
CVS.  See <a href="http://anoncvs.R-project.org">http://anoncvs.R-project.org</a> for more information.
 330 
CVS.  See <a href="http://anoncvs.R-project.org">http://anoncvs.R-project.org</a> for more information.
331 
 
 331 
 
332 
<div class="node">
 332 
<div class="node">
333 
<p><hr>
 333 
<p><hr>
334 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
 334 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
335 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
 335 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
336 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
 336 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
337 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 337 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
338 
<br>
 338 
<br>
339 
</div>
 339 
</div>
340 
 
 340 
 
341 
<h3 class="section">2.5 How can R be installed?</h3>
 341 
<h3 class="section">2.5 How can R be installed?</h3>
342 
 
 342 
 
343 
<ul class="menu">
 343 
<ul class="menu">
344 
<li><a accesskey="1" href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>:
344 
<li><a accesskey="1" href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>:
345 
<li><a accesskey="2" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>:
345 
<li><a accesskey="2" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>:
346 
<li><a accesskey="3" href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>:
346 
<li><a accesskey="3" href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>:
347 
</ul>
 347 
</ul>
348 
 
 348 
 
349 
<div class="node">
 349 
<div class="node">
350 
<p><hr>
 350 
<p><hr>
351 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
 351 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
352 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
 352 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
353 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
 353 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
354 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
 354 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
355 
<br>
 355 
<br>
356 
</div>
 356 
</div>
357 
 
 357 
 
358 
<h3 class="subsection">2.5.1 How can R be installed (Unix)</h4>
 358 
<h3 class="subsection">2.5.1 How can R be installed (Unix)</h4>
359 
 
 359 
 
360 
<p>If binaries are available for your platform (see <a href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>), you can use these, following the instructions that
 360 
<p>If binaries are available for your platform (see <a href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>), you can use these, following the instructions that
361 
come with them.
 361 
come with them.
362 
 
 362 
 
363 
   <p>Otherwise, you can compile and install R yourself, which can be done
 363 
   <p>Otherwise, you can compile and install R yourself, which can be done
364 
very easily under a number of common Unix platforms (see <a href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>).  The file <code>INSTALL</code> that comes with the
 364 
very easily under a number of common Unix platforms (see <a href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>).  The file <code>INSTALL</code> that comes with the
365 
R distribution contains a brief introduction, and the "R Installation
 365 
R distribution contains a brief introduction, and the "R Installation
366 
and Administration" guide (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>)
 366 
and Administration" guide (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>)
367 
has full details.
 367 
has full details.
368 
 
 368 
 
369 
   <p>Note that you need a FORTRAN compiler or <code>f2c</code> in addition to
 369 
   <p>Note that you need a FORTRAN compiler or <code>f2c</code> in addition to
370 
a C compiler to build R.  Also, you need Perl version 5 to build the R
 370 
a C compiler to build R.  Also, you need Perl version 5 to build the R
371 
object documentations.  (If this is not available on your system, you
 371 
object documentations.  (If this is not available on your system, you
372 
can obtain a PDF version of the object reference manual via <small>CRAN</small>.)
 372 
can obtain a PDF version of the object reference manual via <small>CRAN</small>.)
373 
 
 373 
 
374 
   <p>In the simplest case, untar the R source code, change to the directory
 374 
   <p>In the simplest case, untar the R source code, change to the directory
375 
thus created, and issue the following commands (at the shell prompt):
 375 
thus created, and issue the following commands (at the shell prompt):
376 
 
 376 
 
377 
<pre class="example">     $ ./configure
 377 
<pre class="example">     $ ./configure
378 
     $ make
 378 
     $ make
379 
     </pre>
 379 
     </pre>
380 
 
 380 
 
381 
   <p>If these commands execute successfully, the R binary and a shell script
 381 
   <p>If these commands execute successfully, the R binary and a shell script
382 
front-end called <code>R</code> are created and copied to the <code>bin</code>
 382 
front-end called <code>R</code> are created and copied to the <code>bin</code>
383 
directory.  You can copy the script to a place where users can invoke
 383 
directory.  You can copy the script to a place where users can invoke
384 
it, for example to <code>/usr/local/bin</code>.  In addition, plain text help
 384 
it, for example to <code>/usr/local/bin</code>.  In addition, plain text help
385 
pages as well as <small>HTML</small> and LaTeX versions of the documentation are
 385 
pages as well as <small>HTML</small> and LaTeX versions of the documentation are
386 
built.
 386 
built.
387 
 
 387 
 
388 
   <p>Use <kbd>make dvi</kbd> to create DVI versions of the R manuals, such as
 388 
   <p>Use <kbd>make dvi</kbd> to create DVI versions of the R manuals, such as
389 
<code>refman.dvi</code> (an R object reference index) and <code>R-exts.dvi</code>,
 389 
<code>refman.dvi</code> (an R object reference index) and <code>R-exts.dvi</code>,
390 
the "R Extension Writers Guide", in the <code>doc/manual</code>
 390 
the "R Extension Writers Guide", in the <code>doc/manual</code>
391 
subdirectory.  These files can be previewed and printed using standard
 391 
subdirectory.  These files can be previewed and printed using standard
392 
programs such as <code>xdvi</code> and <code>dvips</code>.  You can also use
 392 
programs such as <code>xdvi</code> and <code>dvips</code>.  You can also use
393 
<kbd>make pdf</kbd> to build PDF (Portable Document Format) version of the
 393 
<kbd>make pdf</kbd> to build PDF (Portable Document Format) version of the
394 
manuals, and view these using e.g. Acrobat.  Manuals written in the
 394 
manuals, and view these using e.g. Acrobat.  Manuals written in the
395 
<small>GNU</small> Texinfo system can also be converted to info files
 395 
<small>GNU</small> Texinfo system can also be converted to info files
396 
suitable for reading online with Emacs or stand-alone <small>GNU</small>
 396 
suitable for reading online with Emacs or stand-alone <small>GNU</small>
397 
Info; use <kbd>make info</kbd> to create these versions (note that this
 397 
Info; use <kbd>make info</kbd> to create these versions (note that this
398 
requires <code>makeinfo</code> version 4).
 398 
requires <code>makeinfo</code> version 4).
399 
 
 399 
 
400 
   <p>Finally, use <kbd>make check</kbd> to find out whether your R system works
 400 
   <p>Finally, use <kbd>make check</kbd> to find out whether your R system works
401 
correctly.
 401 
correctly.
402 
 
 402 
 
403 
   <p>You can also perform a "system-wide" installation using <kbd>make
 403 
   <p>You can also perform a "system-wide" installation using <kbd>make
404 
install</kbd>.  By default, this will install to the following directories:
 404 
install</kbd>.  By default, this will install to the following directories:
405 
 
 405 
 
406 
     <dl>
 406 
     <dl>
407 
<dt><code>${prefix}/bin</code>
 407 
<dt><code>${prefix}/bin</code>
408 
     <dd>the front-end shell script
 408 
     <dd>the front-end shell script
409 
<br><dt><code>${prefix}/man/man1</code>
 409 
<br><dt><code>${prefix}/man/man1</code>
410 
     <dd>the man page
 410 
     <dd>the man page
411 
<br><dt><code>${prefix}/lib/R</code>
 411 
<br><dt><code>${prefix}/lib/R</code>
412 
     <dd>all the rest (libraries, on-line help system, <small class="dots">...</small>).  This is the "R
 412 
     <dd>all the rest (libraries, on-line help system, <small class="dots">...</small>).  This is the "R
413 
Home Directory" (<code>R_HOME</code>) of the installed system.
413 
Home Directory" (<code>R_HOME</code>) of the installed system.
414 
</dl>
 414 
</dl>
415 
 
 415 
 
416 
<p>In the above, <code>prefix</code> is determined during configuration
 416 
<p>In the above, <code>prefix</code> is determined during configuration
417 
(typically <code>/usr/local</code>) and can be set by running
 417 
(typically <code>/usr/local</code>) and can be set by running
418 
<code>configure</code> with the option
 418 
<code>configure</code> with the option
419 
 
 419 
 
420 
<pre class="example">     $ ./configure --prefix=/where/you/want/R/to/go
 420 
<pre class="example">     $ ./configure --prefix=/where/you/want/R/to/go
421 
     </pre>
 421 
     </pre>
422 
 
 422 
 
423 
<p>(E.g., the R executable will then be installed into
 423 
<p>(E.g., the R executable will then be installed into
424 
<code>/where/you/want/R/to/go/bin</code>.)
 424 
<code>/where/you/want/R/to/go/bin</code>.)
425 
 
 425 
 
426 
   <p>To install DVI, info and PDF versions of the manuals, use <kbd>make
 426 
   <p>To install DVI, info and PDF versions of the manuals, use <kbd>make
427 
install-dvi</kbd>, <kbd>make install-info</kbd> and <kbd>make install-pdf</kbd>,
 427 
install-dvi</kbd>, <kbd>make install-info</kbd> and <kbd>make install-pdf</kbd>,
428 
respectively.
 428 
respectively.
429 
 
 429 
 
430 
<div class="node">
 430 
<div class="node">
431 
<p><hr>
 431 
<p><hr>
432 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
 432 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
433 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
 433 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
434 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
 434 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
435 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
 435 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
436 
<br>
 436 
<br>
437 
</div>
 437 
</div>
438 
 
 438 
 
439 
<h3 class="subsection">2.5.2 How can R be installed (Windows)</h4>
 439 
<h3 class="subsection">2.5.2 How can R be installed (Windows)</h4>
440 
 
 440 
 
441 
<p>The <code>bin/windows</code> directory of a <small>CRAN</small> site contains binaries for
 441 
<p>The <code>bin/windows</code> directory of a <small>CRAN</small> site contains binaries for
442 
a base distribution and a large number of add-on packages from <small>CRAN</small>
 442 
a base distribution and a large number of add-on packages from <small>CRAN</small>
443 
to run on Windows 95, 98, ME, NT4, 2000, and XP (at least) on Intel and
 443 
to run on Windows 95, 98, ME, NT4, 2000, and XP (at least) on Intel and
444 
clones (but not on other platforms).  The Windows version of R was
 444 
clones (but not on other platforms).  The Windows version of R was
445 
created by Robert Gentleman, and is now being developed and maintained
 445 
created by Robert Gentleman, and is now being developed and maintained
446 
by <a href="mailto:murdoch@stats.uwo.ca">Duncan Murdoch</a> and
 446 
by <a href="mailto:murdoch@stats.uwo.ca">Duncan Murdoch</a> and
447 
<a href="mailto:Brian.Ripley@R-project.org">Brian D. Ripley</a>.
 447 
<a href="mailto:Brian.Ripley@R-project.org">Brian D. Ripley</a>.
448 
 
 448 
 
449 
   <p>For most installations the Windows installer program will be the easiest
 449 
   <p>For most installations the Windows installer program will be the easiest
450 
tool to use.
 450 
tool to use.
451 
 
 451 
 
452 
   <p>See the <a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">"R for Windows <small>FAQ</small>"</a> for more details.
 452 
   <p>See the <a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">"R for Windows <small>FAQ</small>"</a> for more details.
453 
 
 453 
 
454 
<div class="node">
 454 
<div class="node">
455 
<p><hr>
 455 
<p><hr>
456 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
 456 
Node:&nbsp;<a name="How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
457 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
 457 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
458 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
 458 
Up:&nbsp;<a rel="up" accesskey="u" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
459 
<br>
 459 
<br>
460 
</div>
 460 
</div>
461 
 
 461 
 
462 
<h3 class="subsection">2.5.3 How can R be installed (Macintosh)</h4>
 462 
<h3 class="subsection">2.5.3 How can R be installed (Macintosh)</h4>
463 
 
 463 
 
464 
<p>The <code>bin/macosx</code> directory of a <small>CRAN</small> site contains a standard
 464 
<p>The <code>bin/macosx</code> directory of a <small>CRAN</small> site contains a standard
465 
Apple installer package named <code>RAqua.pkg.sit</code> compressed in Aladdin
 465 
Apple installer package named <code>RAqua.pkg.sit</code> compressed in Aladdin
466 
Stuffit format.  Once downloaded, uncompressed and executed, the
 466 
Stuffit format.  Once downloaded, uncompressed and executed, the
467 
installer will install the current non-developer release of R.  RAqua is
 467 
installer will install the current non-developer release of R.  RAqua is
468 
a native MacOSX Darwin version of R with an Aqua GUI.  Inside
 468 
a native MacOSX Darwin version of R with an Aqua GUI.  Inside
469 
<code>bin/macosx/</code><var>x</var><code>.</code><var>y</var><code></code> there are prebuilt binary packages to
 469 
<code>bin/macosx/</code><var>x</var><code>.</code><var>y</var><code></code> there are prebuilt binary packages to
470 
be used with RAqua corresponding to the "<var>x</var>.<var>y</var>" release of
 470 
be used with RAqua corresponding to the "<var>x</var>.<var>y</var>" release of
471 
R. The installation of these packages is available through the
 471 
R. The installation of these packages is available through the
472 
"Package" menu of the RAqua GUI.  This port of R for MacOSX is
 472 
"Package" menu of the RAqua GUI.  This port of R for MacOSX is
473 
maintained by <a href="mailto:Stefano.Iacus@R-project.org">Stefano Iacus</a>.  The
 473 
maintained by <a href="mailto:Stefano.Iacus@R-project.org">Stefano Iacus</a>.  The
474 
<a href="http://cran.r-project.org/bin/macosx/RAqua-FAQ.html">"R for Macintosh <small>FAQ</small>/DOC"</a> has more details.
 474 
<a href="http://cran.r-project.org/bin/macosx/RAqua-FAQ.html">"R for Macintosh <small>FAQ</small>/DOC"</a> has more details.
475 
 
 475 
 
476 
   <p>The <code>bin/macos</code> directory of a <small>CRAN</small> site contains bin-hexed
 476 
   <p>The <code>bin/macos</code> directory of a <small>CRAN</small> site contains bin-hexed
477 
(<code>hqx</code>) and stuffit (<code>sit</code>) archives for a base distribution
 477 
(<code>hqx</code>) and stuffit (<code>sit</code>) archives for a base distribution
478 
and a large number of add-on packages of R 1.7.1 to run under MacOS 8.6
 478 
and a large number of add-on packages of R 1.7.1 to run under MacOS 8.6
479 
to MacOS 9.2.2.  This port of R for Macintosh is no longer supported.
 479 
to MacOS 9.2.2.  This port of R for Macintosh is no longer supported.
480 
 
 480 
 
481 
<div class="node">
 481 
<div class="node">
482 
<p><hr>
 482 
<p><hr>
483 
Node:&nbsp;<a name="Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
 483 
Node:&nbsp;<a name="Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
484 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
 484 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
485 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
 485 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
486 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 486 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
487 
<br>
 487 
<br>
488 
</div>
 488 
</div>
489 
 
 489 
 
490 
<h3 class="section">2.6 Are there Unix binaries for R?</h3>
 490 
<h3 class="section">2.6 Are there Unix binaries for R?</h3>
491 
 
 491 
 
492 
<p>The <code>bin/linux</code> directory of a <small>CRAN</small> site contains Debian
 492 
<p>The <code>bin/linux</code> directory of a <small>CRAN</small> site contains Debian
493 
stable/testing/unstable packages for the i386 platform (now part of the
 493 
stable/testing/unstable packages for the i386 platform (now part of the
494 
Debian distribution and maintained by Dirk Eddelbuettel), Mandrake
 494 
Debian distribution and maintained by Dirk Eddelbuettel), Mandrake
495 
9.0/9.1 i386 packages by Michele Alzetta, Red Hat 7.x/8.x/9 i386
 495 
9.0/9.1 i386 packages by Michele Alzetta, Red Hat 7.x/8.x/9 i386
496 
packages by Martyn Plummer, SuSE 7.3/8.0/8.1/8.2/9.0 i386 packages by
 496 
packages by Martyn Plummer, SuSE 7.3/8.0/8.1/8.2/9.0 i386 packages by
497 
Detlef Steuer, and VineLinux 2.6 i386 packages by Susunu Tanimura.
 497 
Detlef Steuer, and VineLinux 2.6 i386 packages by Susunu Tanimura.
498 
 
 498 
 
499 
   <p>The Debian packages can be accessed through APT, the Debian package
 499 
   <p>The Debian packages can be accessed through APT, the Debian package
500 
maintenance tool.  Simply add the line
 500 
maintenance tool.  Simply add the line
501 
 
 501 
 
502 
<pre class="example">     deb http://cran.R-project.org/bin/linux/debian <var>distribution</var> main
 502 
<pre class="example">     deb http://cran.R-project.org/bin/linux/debian <var>distribution</var> main
503 
     </pre>
 503 
     </pre>
504 
 
 504 
 
505 
<p>(where <var>distribution</var> is either <code>stable</code> or <code>testing</code>;
 505 
<p>(where <var>distribution</var> is either <code>stable</code> or <code>testing</code>;
506 
feel free to use a <small>CRAN</small> mirror instead of the master) to the file
 506 
feel free to use a <small>CRAN</small> mirror instead of the master) to the file
507 
<code>/etc/apt/sources.list</code>.  Once you have added that line the
 507 
<code>/etc/apt/sources.list</code>.  Once you have added that line the
508 
programs <code>apt-get</code>, <code>apt-cache</code>, and <code>dselect</code>
 508 
programs <code>apt-get</code>, <code>apt-cache</code>, and <code>dselect</code>
509 
(using the apt access method) will automatically detect and install
 509 
(using the apt access method) will automatically detect and install
510 
updates of the R packages.
 510 
updates of the R packages.
511 
 
 511 
 
512 
   <p>No other binary distributions are currently publically available.
 512 
   <p>No other binary distributions are currently publically available.
513 
 
 513 
 
514 
<div class="node">
 514 
<div class="node">
515 
<p><hr>
 515 
<p><hr>
516 
Node:&nbsp;<a name="What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
 516 
Node:&nbsp;<a name="What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
517 
Next:&nbsp;<a rel="next" accesskey="n" href="#Citing%20R">Citing R</a>,
 517 
Next:&nbsp;<a rel="next" accesskey="n" href="#Citing%20R">Citing R</a>,
518 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
 518 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
519 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 519 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
520 
<br>
 520 
<br>
521 
</div>
 521 
</div>
522 
 
 522 
 
523 
<h3 class="section">2.7 What documentation exists for R?</h3>
 523 
<h3 class="section">2.7 What documentation exists for R?</h3>
524 
 
 524 
 
525 
<p>Online documentation for most of the functions and variables in R
 525 
<p>Online documentation for most of the functions and variables in R
526 
exists, and can be printed on-screen by typing <kbd>help(</kbd><var>name</var><kbd>)</kbd>
 526 
exists, and can be printed on-screen by typing <kbd>help(</kbd><var>name</var><kbd>)</kbd>
527 
(or <kbd>?</kbd><var>name</var><kbd></kbd>) at the R prompt, where <var>name</var> is the name of
 527 
(or <kbd>?</kbd><var>name</var><kbd></kbd>) at the R prompt, where <var>name</var> is the name of
528 
the topic help is sought for.  (In the case of unary and binary
 528 
the topic help is sought for.  (In the case of unary and binary
529 
operators and control-flow special forms, the name may need to be be
 529 
operators and control-flow special forms, the name may need to be be
530 
quoted.)
 530 
quoted.)
531 
 
 531 
 
532 
   <p>This documentation can also be made available as one reference manual
 532 
   <p>This documentation can also be made available as one reference manual
533 
for on-line reading in <small>HTML</small> and PDF formats, and as hardcopy via
 533 
for on-line reading in <small>HTML</small> and PDF formats, and as hardcopy via
534 
LaTeX, see <a href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>.  An up-to-date <small>HTML</small>
 534 
LaTeX, see <a href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>.  An up-to-date <small>HTML</small>
535 
version is always available for web browsing at
 535 
version is always available for web browsing at
536 
<a href="http://stat.ethz.ch/R-manual/">http://stat.ethz.ch/R-manual/</a>.
 536 
<a href="http://stat.ethz.ch/R-manual/">http://stat.ethz.ch/R-manual/</a>.
537 
 
 537 
 
538 
   <p>The R distribution also comes with the following manuals.
 538 
   <p>The R distribution also comes with the following manuals.
539 
 
 539 
 
540 
     <ul>
 540 
     <ul>
541 
<li>"An Introduction to R" (<code>R-intro</code>)
 541 
<li>"An Introduction to R" (<code>R-intro</code>)
542 
includes information on data types, programming elements, statistical
 542 
includes information on data types, programming elements, statistical
543 
modeling and graphics.  This document is based on the "Notes on
 543 
modeling and graphics.  This document is based on the "Notes on
544 
<small>S-PLUS</small>" by Bill Venables and David Smith.
544 
<small>S-PLUS</small>" by Bill Venables and David Smith.
545 
<li>"Writing R Extensions" (<code>R-exts</code>)
 545 
<li>"Writing R Extensions" (<code>R-exts</code>)
546 
currently describes the process of creating R add-on packages, writing R
 546 
currently describes the process of creating R add-on packages, writing R
547 
documentation, R's system and foreign language interfaces, and the R
 547 
documentation, R's system and foreign language interfaces, and the R
548 
<small>API</small>.
548 
<small>API</small>.
549 
<li>"R Data Import/Export" (<code>R-data</code>)
 549 
<li>"R Data Import/Export" (<code>R-data</code>)
550 
is a guide to importing and exporting data to and from R.
550 
is a guide to importing and exporting data to and from R.
551 
<li>"The R Language Definition" (<code>R-lang</code>),
 551 
<li>"The R Language Definition" (<code>R-lang</code>),
552 
a first version of the "Kernighan &amp; Ritchie of R", explains
 552 
a first version of the "Kernighan &amp; Ritchie of R", explains
553 
evaluation, parsing, object oriented programming, computing on the
 553 
evaluation, parsing, object oriented programming, computing on the
554 
language, and so forth.
554 
language, and so forth.
555 
<li>"R Installation and Administration" (<code>R-admin</code>).
555 
<li>"R Installation and Administration" (<code>R-admin</code>).
556 
</ul>
 556 
</ul>
557 
 
 557 
 
558 
   <p>Books on R include
 558 
   <p>Books on R include
559 
 
 559 
 
560 
   <blockquote>
 560 
   <blockquote>
561 
P. Dalgaard (2002), "Introductory Statistics with R", Springer: New
 561 
P. Dalgaard (2002), "Introductory Statistics with R", Springer: New
562 
York, ISBN 0-387-95475-9.
 562 
York, ISBN 0-387-95475-9.
563 
 
 563 
 
564 
        <p>J. Fox (2002), "An R and <small>S-PLUS</small> Companion to Applied Regression",
 564 
        <p>J. Fox (2002), "An R and <small>S-PLUS</small> Companion to Applied Regression",
565 
Sage Publications, ISBN 0-761-92280-6 (softcover) or 0-761-92279-2
 565 
Sage Publications, ISBN 0-761-92280-6 (softcover) or 0-761-92279-2
566 
(hardcover),
 566 
(hardcover),
567 
<a href="http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/">http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/</a>.
 567 
<a href="http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/">http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/</a>.
568 
 
 568 
 
569 
        <p>J. Maindonald and J. Braun (2003), "Data Analysis and Graphics Using R:
 569 
        <p>J. Maindonald and J. Braun (2003), "Data Analysis and Graphics Using R:
570 
An Example-Based Approach", Cambridge University Press, ISBN
 570 
An Example-Based Approach", Cambridge University Press, ISBN
571 
0-521-81336-0, <a href="http://wwwmaths.anu.edu.au/~johnm/">http://wwwmaths.anu.edu.au/~johnm/</a>.
 571 
0-521-81336-0, <a href="http://wwwmaths.anu.edu.au/~johnm/">http://wwwmaths.anu.edu.au/~johnm/</a>.
572 
 
 572 
 
573 
        <p>S. M. Iacus and G. Masarotto (2002), "Laboratorio di statistica con R
 573 
        <p>S. M. Iacus and G. Masarotto (2002), "Laboratorio di statistica con R
574 
", McGraw-Hill, ISBN 88-386-6084-0 (in Italian).
574 
", McGraw-Hill, ISBN 88-386-6084-0 (in Italian).
575 
</blockquote>
 575 
</blockquote>
576 
 
 576 
 
577 
<p>The book
 577 
<p>The book
578 
 
 578 
 
579 
   <blockquote>
 579 
   <blockquote>
580 
W. N. Venables and B. D. Ripley (2002), "Modern Applied Statistics with
 580 
W. N. Venables and B. D. Ripley (2002), "Modern Applied Statistics with
581 
S.  Fourth Edition".  Springer, ISBN 0-387-95457-0
 581 
S.  Fourth Edition".  Springer, ISBN 0-387-95457-0
582 
</blockquote>
 582 
</blockquote>
583 
 
 583 
 
584 
<p>has a home page at <a href="http://www.stats.ox.ac.uk/pub/MASS4/">http://www.stats.ox.ac.uk/pub/MASS4/</a> providing
 584 
<p>has a home page at <a href="http://www.stats.ox.ac.uk/pub/MASS4/">http://www.stats.ox.ac.uk/pub/MASS4/</a> providing
585 
additional material.  Its companion is
 585 
additional material.  Its companion is
586 
 
 586 
 
587 
   <blockquote>
 587 
   <blockquote>
588 
W. N. Venables and B. D. Ripley (2000), "S Programming".  Springer,
 588 
W. N. Venables and B. D. Ripley (2000), "S Programming".  Springer,
589 
ISBN 0-387-98966-8
 589 
ISBN 0-387-98966-8
590 
</blockquote>
 590 
</blockquote>
591 
 
 591 
 
592 
<p>and provides an in-depth guide to writing software in the S language
 592 
<p>and provides an in-depth guide to writing software in the S language
593 
which forms the basis of both the commercial <small>S-PLUS</small> and the Open
 593 
which forms the basis of both the commercial <small>S-PLUS</small> and the Open
594 
Source R data analysis software systems.  See
 594 
Source R data analysis software systems.  See
595 
<a href="http://www.stats.ox.ac.uk/pub/MASS3/Sprog/">http://www.stats.ox.ac.uk/pub/MASS3/Sprog/</a> for more information.
 595 
<a href="http://www.stats.ox.ac.uk/pub/MASS3/Sprog/">http://www.stats.ox.ac.uk/pub/MASS3/Sprog/</a> for more information.
596 
 
 596 
 
597 
   <p>In addition to material written specifically or explicitly for R,
 597 
   <p>In addition to material written specifically or explicitly for R,
598 
documentation for S/<small>S-PLUS</small> (see <a href="#R%20and%20S">R and S</a>) can be used in
 598 
documentation for S/<small>S-PLUS</small> (see <a href="#R%20and%20S">R and S</a>) can be used in
599 
combination with this <small>FAQ</small> (see <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>).  Introductory books include
 599 
combination with this <small>FAQ</small> (see <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>).  Introductory books include
600 
 
 600 
 
601 
   <blockquote>
 601 
   <blockquote>
602 
P. Spector (1994), "An introduction to S and <small>S-PLUS</small>", Duxbury Press.
 602 
P. Spector (1994), "An introduction to S and <small>S-PLUS</small>", Duxbury Press.
603 
 
 603 
 
604 
        <p>A. Krause and M. Olsen (2002), "The Basics of <small>S-PLUS</small>" (Third
 604 
        <p>A. Krause and M. Olsen (2002), "The Basics of <small>S-PLUS</small>" (Third
605 
Edition).  Springer, ISBN 0-387-95456-2
 605 
Edition).  Springer, ISBN 0-387-95456-2
606 
</blockquote>
 606 
</blockquote>
607 
 
 607 
 
608 
   <p>The book
 608 
   <p>The book
609 
 
 609 
 
610 
   <blockquote>
 610 
   <blockquote>
611 
J. C. Pinheiro and D. M. Bates (2000), "Mixed-Effects Models in S and
 611 
J. C. Pinheiro and D. M. Bates (2000), "Mixed-Effects Models in S and
612 
<small>S-PLUS</small>", Springer, ISBN 0-387-98957-0
 612 
<small>S-PLUS</small>", Springer, ISBN 0-387-98957-0
613 
</blockquote>
 613 
</blockquote>
614 
 
 614 
 
615 
<p>provides a comprehensive guide to the use of the <strong>nlme</strong> package
 615 
<p>provides a comprehensive guide to the use of the <strong>nlme</strong> package
616 
for linear and nonlinear mixed-effects models.  This has a home page at
 616 
for linear and nonlinear mixed-effects models.  This has a home page at
617 
<a href="http://nlme.stat.wisc.edu/MEMSS/">http://nlme.stat.wisc.edu/MEMSS/</a>.
 617 
<a href="http://nlme.stat.wisc.edu/MEMSS/">http://nlme.stat.wisc.edu/MEMSS/</a>.
618 
 
 618 
 
619 
   <p>As an example of how R can be used in teaching an advanced introductory
 619 
   <p>As an example of how R can be used in teaching an advanced introductory
620 
statistics course, see
 620 
statistics course, see
621 
 
 621 
 
622 
   <blockquote>
 622 
   <blockquote>
623 
D. Nolan and T. Speed (2000), "Stat Labs: Mathematical Statistics
 623 
D. Nolan and T. Speed (2000), "Stat Labs: Mathematical Statistics
624 
Through Applications", Springer Texts in Statistics, ISBN
 624 
Through Applications", Springer Texts in Statistics, ISBN
625 
0-387-98974-9
 625 
0-387-98974-9
626 
</blockquote>
 626 
</blockquote>
627 
 
 627 
 
628 
<p>This integrates theory of statistics with the practice of statistics
 628 
<p>This integrates theory of statistics with the practice of statistics
629 
through a collection of case studies ("labs"), and uses R to analyze
 629 
through a collection of case studies ("labs"), and uses R to analyze
630 
the data.  More information can be found at
 630 
the data.  More information can be found at
631 
<a href="http://www.stat.Berkeley.EDU/users/statlabs/">http://www.stat.Berkeley.EDU/users/statlabs/</a>.
 631 
<a href="http://www.stat.Berkeley.EDU/users/statlabs/">http://www.stat.Berkeley.EDU/users/statlabs/</a>.
632 
 
 632 
 
633 
   <p>Last, but not least, Ross' and Robert's experience in designing and
 633 
   <p>Last, but not least, Ross' and Robert's experience in designing and
634 
implementing R is described in Ihaka &amp; Gentleman (1996), "R: A Language
 634 
implementing R is described in Ihaka &amp; Gentleman (1996), "R: A Language
635 
for Data Analysis and Graphics",
 635 
for Data Analysis and Graphics",
636 
<a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>5</strong>, 299-314.
 636 
<a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>5</strong>, 299-314.
637 
 
 637 
 
638 
   <p>An annotated bibliography (BibTeX format) of R-related publications
 638 
   <p>An annotated bibliography (BibTeX format) of R-related publications
639 
which includes most of the above references can be found at
 639 
which includes most of the above references can be found at
640 
 
 640 
 
641 
<pre class="display">     <a href="http://www.R-project.org/doc/bib/R.bib">http://www.R-project.org/doc/bib/R.bib</a>
 641 
<pre class="display">     <a href="http://www.R-project.org/doc/bib/R.bib">http://www.R-project.org/doc/bib/R.bib</a>
642 
     </pre>
 642 
     </pre>
643 
 
 643 
 
644 
<div class="node">
 644 
<div class="node">
645 
<p><hr>
 645 
<p><hr>
646 
Node:&nbsp;<a name="Citing%20R">Citing R</a>,
 646 
Node:&nbsp;<a name="Citing%20R">Citing R</a>,
647 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
 647 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
648 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
 648 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
649 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 649 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
650 
<br>
 650 
<br>
651 
</div>
 651 
</div>
652 
 
 652 
 
653 
<h3 class="section">2.8 Citing R</h3>
 653 
<h3 class="section">2.8 Citing R</h3>
654 
 
 654 
 
655 
<p>To cite R in publications, use
 655 
<p>To cite R in publications, use
656 
 
 656 
 
657 
<pre class="example">     @Manual{,
 657 
<pre class="example">     @Manual{,
658 
       title        = {R: A language and environment for statistical
 658 
       title        = {R: A language and environment for statistical
659 
                       computing},
 659 
                       computing},
660 
       author       = {{R Development Core Team}},
 660 
       author       = {{R Development Core Team}},
661 
       organization = {R Foundation for Statistical Computing},
 661 
       organization = {R Foundation for Statistical Computing},
662 
       address      = {Vienna, Austria},
 662 
       address      = {Vienna, Austria},
663 
       year         = 2003,
 663 
       year         = 2003,
664 
       note         = {ISBN 3-900051-00-3},
 664 
       note         = {ISBN 3-900051-00-3},
665 
       url          = {http://www.R-project.org}
 665 
       url          = {http://www.R-project.org}
666 
     }
 666 
     }
667 
     </pre>
 667 
     </pre>
668 
 
 668 
 
669 
<div class="node">
 669 
<div class="node">
670 
<p><hr>
 670 
<p><hr>
671 
Node:&nbsp;<a name="What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
 671 
Node:&nbsp;<a name="What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
672 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20CRAN%3f">What is CRAN?</a>,
 672 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20CRAN%3f">What is CRAN?</a>,
673 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Citing%20R">Citing R</a>,
 673 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Citing%20R">Citing R</a>,
674 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 674 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
675 
<br>
 675 
<br>
676 
</div>
 676 
</div>
677 
 
 677 
 
678 
<h3 class="section">2.9 What mailing lists exist for R?</h3>
 678 
<h3 class="section">2.9 What mailing lists exist for R?</h3>
679 
 
 679 
 
680 
<p>Thanks to <a href="mailto:Martin.Maechler@R-project.org">Martin Maechler</a>, there
 680 
<p>Thanks to <a href="mailto:Martin.Maechler@R-project.org">Martin Maechler</a>, there
681 
are four mailing lists devoted to R.
 681 
are four mailing lists devoted to R.
682 
 
 682 
 
683 
     <dl>
 683 
     <dl>
684 
<dt><code>R-announce</code>
 684 
<dt><code>R-announce</code>
685 
     <dd>A moderated list for announcements about the development of R and the
 685 
     <dd>A moderated list for announcements about the development of R and the
686 
availability of new code.
686 
availability of new code.
687 
<br><dt><code>R-packages</code>
 687 
<br><dt><code>R-packages</code>
688 
     <dd>A moderated list for announcements on the availability of new or
 688 
     <dd>A moderated list for announcements on the availability of new or
689 
enhanced contributed packages.
689 
enhanced contributed packages.
690 
<br><dt><code>R-help</code>
 690 
<br><dt><code>R-help</code>
691 
     <dd>The `main' R mailing list, for discussion about problems and solutions
 691 
     <dd>The `main' R mailing list, for discussion about problems and solutions
692 
using R, announcements (not covered by `R-announce' and `R-packages')
 692 
using R, announcements (not covered by `R-announce' and `R-packages')
693 
about the development of R and the availability of new code,
 693 
about the development of R and the availability of new code,
694 
enhancements and patches to the source code and documentation of R,
 694 
enhancements and patches to the source code and documentation of R,
695 
comparison and compatibility with S and <small>S-PLUS</small>, and for the posting of
 695 
comparison and compatibility with S and <small>S-PLUS</small>, and for the posting of
696 
nice examples and benchmarks.
696 
nice examples and benchmarks.
697 
<br><dt><code>R-devel</code>
 697 
<br><dt><code>R-devel</code>
698 
     <dd>This list is for discussions about the future of R and pre-testing of
 698 
     <dd>This list is for discussions about the future of R and pre-testing of
699 
new versions.  It is meant for those who maintain an active position in
 699 
new versions.  It is meant for those who maintain an active position in
700 
the development of R.
700 
the development of R.
701 
</dl>
 701 
</dl>
702 
 
 702 
 
703 
<p>Note that the R-announce and R-packages lists are gatewayed into R-help.
703 
<p>Note that the R-announce and R-packages lists are gatewayed into R-help.
704 
Hence, you should subscribe to either of them only in case you are not
 704 
Hence, you should subscribe to either of them only in case you are not
705 
subscribed to R-help.
 705 
subscribed to R-help.
706 
 
 706 
 
707 
   <p>Send email to <a href="mailto:R-help@lists.R-project.org">R-help@lists.R-project.org</a> to reach everyone on
 707 
   <p>Send email to <a href="mailto:R-help@lists.R-project.org">R-help@lists.R-project.org</a> to reach everyone on
708 
the R-help mailing list.  To subscribe (or unsubscribe) to this list
 708 
the R-help mailing list.  To subscribe (or unsubscribe) to this list
709 
send <code>subscribe</code> (or <code>unsubscribe</code>) in the <em>body</em> of the
 709 
send <code>subscribe</code> (or <code>unsubscribe</code>) in the <em>body</em> of the
710 
message (not in the subject!) to
 710 
message (not in the subject!) to
711 
<a href="mailto:R-help-request@lists.R-project.org">R-help-request@lists.R-project.org</a>.  Information about the list
 711 
<a href="mailto:R-help-request@lists.R-project.org">R-help-request@lists.R-project.org</a>.  Information about the list
712 
can be obtained by sending an email with <code>info</code> as its contents to
 712 
can be obtained by sending an email with <code>info</code> as its contents to
713 
<a href="mailto:R-help-request@lists.R-project.org">R-help-request@lists.R-project.org</a>.
 713 
<a href="mailto:R-help-request@lists.R-project.org">R-help-request@lists.R-project.org</a>.
714 
 
 714 
 
715 
   <p>Subscription and posting to the other lists is done analogously, with
 715 
   <p>Subscription and posting to the other lists is done analogously, with
716 
<code>R-help</code> replaced by <code>R-announce</code>, <code>R-packages</code>, and
 716 
<code>R-help</code> replaced by <code>R-announce</code>, <code>R-packages</code>, and
717 
<code>R-devel</code>, respectively.
 717 
<code>R-devel</code>, respectively.
718 
 
 718 
 
719 
   <p>Subscriptions to the R-help and R-devel mailing lists are also available
 719 
   <p>Subscriptions to the R-help and R-devel mailing lists are also available
720 
in digest (plain or MIME) format, see the <code>doc/html/mail.html</code> file
 720 
in digest (plain or MIME) format, see the <code>doc/html/mail.html</code> file
721 
in <small>CRAN</small> for more information.
 721 
in <small>CRAN</small> for more information.
722 
 
 722 
 
723 
   <p>It is recommended that you send mail to R-help rather than only to the R
 723 
   <p>It is recommended that you send mail to R-help rather than only to the R
724 
Core developers (who are also subscribed to the list, of course).  This
 724 
Core developers (who are also subscribed to the list, of course).  This
725 
may save them precious time they can use for constantly improving R, and
 725 
may save them precious time they can use for constantly improving R, and
726 
will typically also result in much quicker feedback for yourself.
 726 
will typically also result in much quicker feedback for yourself.
727 
 
 727 
 
728 
   <p>Of course, in the case of bug reports it would be very helpful to have
 728 
   <p>Of course, in the case of bug reports it would be very helpful to have
729 
code which reliably reproduces the problem.  Also, make sure that you
 729 
code which reliably reproduces the problem.  Also, make sure that you
730 
include information on the system and version of R being used.  See
 730 
include information on the system and version of R being used.  See
731 
<a href="#R%20Bugs">R Bugs</a> for more details.
 731 
<a href="#R%20Bugs">R Bugs</a> for more details.
732 
 
 732 
 
733 
   <p>Archives of the above three mailing lists are made available on the net
 733 
   <p>Archives of the above three mailing lists are made available on the net
734 
in a monthly schedule via the <code>doc/html/mail.html</code> file in <small>CRAN</small>.
734 
in a monthly schedule via the <code>doc/html/mail.html</code> file in <small>CRAN</small>.
735 
Searchable archives of the lists are available via
 735 
Searchable archives of the lists are available via
736 
<a href="http://maths.newcastle.edu.au/~rking/R/">http://maths.newcastle.edu.au/~rking/R/</a>.
 736 
<a href="http://maths.newcastle.edu.au/~rking/R/">http://maths.newcastle.edu.au/~rking/R/</a>.
737 
 
 737 
 
738 
   <p>The R Core Team can be reached at <a href="mailto:R-core@lists.R-project.org">R-core@lists.R-project.org</a>
 738 
   <p>The R Core Team can be reached at <a href="mailto:R-core@lists.R-project.org">R-core@lists.R-project.org</a>
739 
for comments and reports.
 739 
for comments and reports.
740 
 
 740 
 
741 
<div class="node">
 741 
<div class="node">
742 
<p><hr>
 742 
<p><hr>
743 
Node:&nbsp;<a name="What%20is%20CRAN%3f">What is CRAN?</a>,
 743 
Node:&nbsp;<a name="What%20is%20CRAN%3f">What is CRAN?</a>,
744 
Next:&nbsp;<a rel="next" accesskey="n" href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
 744 
Next:&nbsp;<a rel="next" accesskey="n" href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
745 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
 745 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
746 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 746 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
747 
<br>
 747 
<br>
748 
</div>
 748 
</div>
749 
 
 749 
 
750 
<h3 class="section">2.10 What is <small>CRAN</small>?</h3>
 750 
<h3 class="section">2.10 What is <small>CRAN</small>?</h3>
751 
 
 751 
 
752 
<p>The "Comprehensive R Archive Network" (<small>CRAN</small>) is a collection of
 752 
<p>The "Comprehensive R Archive Network" (<small>CRAN</small>) is a collection of
753 
sites which carry identical material, consisting of the R
 753 
sites which carry identical material, consisting of the R
754 
distribution(s), the contributed extensions, documentation for R, and
 754 
distribution(s), the contributed extensions, documentation for R, and
755 
binaries.
 755 
binaries.
756 
 
 756 
 
757 
   <p>The <small>CRAN</small> master site at TU Wien, Austria, can be found at the
 757 
   <p>The <small>CRAN</small> master site at TU Wien, Austria, can be found at the
758 
<small>URL</small>
 758 
<small>URL</small>
759 
 
 759 
 
760 
   <blockquote>
 760 
   <blockquote>
761 
<a href="http://cran.R-project.org/">http://cran.R-project.org/</a>
 761 
<a href="http://cran.R-project.org/">http://cran.R-project.org/</a>
762 
</blockquote>
 762 
</blockquote>
763 
 
 763 
 
764 
<p>and is currently being mirrored daily at
 764 
<p>and is currently being mirrored daily at
765 
 
 765 
 
766 
   <blockquote>
 766 
   <blockquote>
767 
        <p><table><tr align="left"><td valign="top"><a href="http://cran.at.R-project.org/">http://cran.at.R-project.org/</a>
 767 
        <p><table><tr align="left"><td valign="top"><a href="http://cran.at.R-project.org/">http://cran.at.R-project.org/</a>
768 
</td><td valign="top">(TU Wien, Austria)
 768 
</td><td valign="top">(TU Wien, Austria)
769 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.au.R-project.org/">http://cran.au.R-project.org/</a>
 769 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.au.R-project.org/">http://cran.au.R-project.org/</a>
770 
</td><td valign="top">(PlanetMirror, Australia)
 770 
</td><td valign="top">(PlanetMirror, Australia)
771 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.br.R-project.org/">http://cran.br.R-project.org/</a>
 771 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.br.R-project.org/">http://cran.br.R-project.org/</a>
772 
</td><td valign="top">(Universidade Federal de Paran&aacute;, Brazil)
 772 
</td><td valign="top">(Universidade Federal de Paran&aacute;, Brazil)
773 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.ch.R-project.org/">http://cran.ch.R-project.org/</a>
 773 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.ch.R-project.org/">http://cran.ch.R-project.org/</a>
774 
</td><td valign="top">(ETH Z&uuml;rich, Switzerland)
 774 
</td><td valign="top">(ETH Z&uuml;rich, Switzerland)
775 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.de.R-project.org/">http://cran.de.R-project.org/</a>
 775 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.de.R-project.org/">http://cran.de.R-project.org/</a>
776 
</td><td valign="top">(APP, Germany)
 776 
</td><td valign="top">(APP, Germany)
777 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.dk.R-project.org/">http://cran.dk.R-project.org/</a>
 777 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.dk.R-project.org/">http://cran.dk.R-project.org/</a>
778 
</td><td valign="top">(SunSITE, Denmark)
 778 
</td><td valign="top">(SunSITE, Denmark)
779 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.es.R-project.org/">http://cran.es.R-project.org/</a>
 779 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.es.R-project.org/">http://cran.es.R-project.org/</a>
780 
</td><td valign="top">(Spanish National Research Network, Madrid, Spain)
 780 
</td><td valign="top">(Spanish National Research Network, Madrid, Spain)
781 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.hu.R-project.org/">http://cran.hu.R-project.org/</a>
 781 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.hu.R-project.org/">http://cran.hu.R-project.org/</a>
782 
</td><td valign="top">(Semmelweis U, Hungary)
 782 
</td><td valign="top">(Semmelweis U, Hungary)
783 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.uk.R-project.org/">http://cran.uk.R-project.org/</a>
 783 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.uk.R-project.org/">http://cran.uk.R-project.org/</a>
784 
</td><td valign="top">(U of Bristol, United Kingdom)
 784 
</td><td valign="top">(U of Bristol, United Kingdom)
785 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.us.R-project.org/">http://cran.us.R-project.org/</a>
 785 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.us.R-project.org/">http://cran.us.R-project.org/</a>
786 
</td><td valign="top">(U of Wisconsin, USA)
 786 
</td><td valign="top">(U of Wisconsin, USA)
787 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.za.R-project.org/">http://cran.za.R-project.org/</a>
 787 
<br></td></tr><tr align="left"><td valign="top"><a href="http://cran.za.R-project.org/">http://cran.za.R-project.org/</a>
788 
</td><td valign="top">(Rhodes U, South Africa)
 788 
</td><td valign="top">(Rhodes U, South Africa)
789 
        <br></td></tr></table>
 789 
        <br></td></tr></table>
790 
</blockquote>
 790 
</blockquote>
791 
 
 791 
 
792 
<p>Please use the <small>CRAN</small> site closest to you to reduce network load.
 792 
<p>Please use the <small>CRAN</small> site closest to you to reduce network load.
793 
 
 793 
 
794 
   <p>From <small>CRAN</small>, you can obtain the latest official release of R, daily
 794 
   <p>From <small>CRAN</small>, you can obtain the latest official release of R, daily
795 
snapshots of R (copies of the current CVS trees), as gzipped and bzipped
 795 
snapshots of R (copies of the current CVS trees), as gzipped and bzipped
796 
tar files, a wealth of additional contributed code, as well as prebuilt
 796 
tar files, a wealth of additional contributed code, as well as prebuilt
797 
binaries for various operating systems (Linux, MacOS Classic, MacOS X,
 797 
binaries for various operating systems (Linux, MacOS Classic, MacOS X,
798 
and MS Windows).  <small>CRAN</small> also provides access to documentation on R,
 798 
and MS Windows).  <small>CRAN</small> also provides access to documentation on R,
799 
existing mailing lists and the R Bug Tracking system.
 799 
existing mailing lists and the R Bug Tracking system.
800 
 
 800 
 
801 
   <p>To "submit" to <small>CRAN</small>, simply upload to
 801 
   <p>To "submit" to <small>CRAN</small>, simply upload to
802 
<a href="ftp://cran.R-project.org/incoming/">ftp://cran.R-project.org/incoming/</a> and send an email to
 802 
<a href="ftp://cran.R-project.org/incoming/">ftp://cran.R-project.org/incoming/</a> and send an email to
803 
<a href="mailto:cran@R-project.org">cran@R-project.org</a>.  Note that <small>CRAN</small> generally does not
 803 
<a href="mailto:cran@R-project.org">cran@R-project.org</a>.  Note that <small>CRAN</small> generally does not
804 
accept submissions of precompiled binaries due to security reasons.
 804 
accept submissions of precompiled binaries due to security reasons.
805 
 
 805 
 
806 
   <blockquote>
 806 
   <blockquote>
807 
<strong>Note:</strong>  It is very important that you indicate the copyright
 807 
<strong>Note:</strong>  It is very important that you indicate the copyright
808 
(license) information (<small>GPL</small>, <small>BSD</small>, Artistic, <small class="dots">...</small>)
 808 
(license) information (<small>GPL</small>, <small>BSD</small>, Artistic, <small class="dots">...</small>)
809 
in your submission.
809 
in your submission.
810 
</blockquote>
 810 
</blockquote>
811 
 
 811 
 
812 
   <p>Please always use the <small>URL</small> of the master site when referring to
 812 
   <p>Please always use the <small>URL</small> of the master site when referring to
813 
<small>CRAN</small>.
 813 
<small>CRAN</small>.
814 
 
 814 
 
815 
<div class="node">
 815 
<div class="node">
816 
<p><hr>
 816 
<p><hr>
817 
Node:&nbsp;<a name="Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
 817 
Node:&nbsp;<a name="Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
818 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20is%20R%20named%20R%3f">Why is R named R?</a>,
 818 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20is%20R%20named%20R%3f">Why is R named R?</a>,
819 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20CRAN%3f">What is CRAN?</a>,
 819 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20CRAN%3f">What is CRAN?</a>,
820 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 820 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
821 
<br>
 821 
<br>
822 
</div>
 822 
</div>
823 
 
 823 
 
824 
<h3 class="section">2.11 Can I use R for commercial purposes?</h3>
 824 
<h3 class="section">2.11 Can I use R for commercial purposes?</h3>
825 
 
 825 
 
826 
<p>R is released under the <a href="http://www.gnu.org/copyleft/gpl.html">GNU General Public License (GPL)</a>.  If you have any questions regarding the
 826 
<p>R is released under the <a href="http://www.gnu.org/copyleft/gpl.html">GNU General Public License (GPL)</a>.  If you have any questions regarding the
827 
legality of using R in any particular situation you should bring it up
 827 
legality of using R in any particular situation you should bring it up
828 
with your legal counsel.  We are in no position to offer legal advice.
 828 
with your legal counsel.  We are in no position to offer legal advice.
829 
 
 829 
 
830 
   <p>It is the opinion of the R Core Team that one can use R for commercial
 830 
   <p>It is the opinion of the R Core Team that one can use R for commercial
831 
purposes (e.g., in business or in consulting).  The GPL, like all Open
 831 
purposes (e.g., in business or in consulting).  The GPL, like all Open
832 
Source licenses, permits all and any use of the package.  It only
 832 
Source licenses, permits all and any use of the package.  It only
833 
restricts distribution of R or of other programs containing code from R.
833 
restricts distribution of R or of other programs containing code from R.
834 
This is made clear in clause 6 ("No Discrimination Against Fields of
 834 
This is made clear in clause 6 ("No Discrimination Against Fields of
835 
Endeavor") of the <a href="http://www.opensource.org/docs/definition.html">Open Source Definition</a>:
 835 
Endeavor") of the <a href="http://www.opensource.org/docs/definition.html">Open Source Definition</a>:
836 
 
 836 
 
837 
   <blockquote>
 837 
   <blockquote>
838 
The license must not restrict anyone from making use of the program in a
 838 
The license must not restrict anyone from making use of the program in a
839 
specific field of endeavor.  For example, it may not restrict the
 839 
specific field of endeavor.  For example, it may not restrict the
840 
program from being used in a business, or from being used for genetic
 840 
program from being used in a business, or from being used for genetic
841 
research.
841 
research.
842 
</blockquote>
 842 
</blockquote>
843 
 
 843 
 
844 
<p>It is also explicitly stated in clause 0 of the GPL, which says in part
 844 
<p>It is also explicitly stated in clause 0 of the GPL, which says in part
845 
 
 845 
 
846 
   <blockquote>
 846 
   <blockquote>
847 
Activities other than copying, distribution and modification are not
 847 
Activities other than copying, distribution and modification are not
848 
covered by this License; they are outside its scope.  The act of running
 848 
covered by this License; they are outside its scope.  The act of running
849 
the Program is not restricted, and the output from the Program is
 849 
the Program is not restricted, and the output from the Program is
850 
covered only if its contents constitute a work based on the Program.
850 
covered only if its contents constitute a work based on the Program.
851 
</blockquote>
 851 
</blockquote>
852 
 
 852 
 
853 
   <p>Most add-on packages, including all recommended ones, also explicitly
 853 
   <p>Most add-on packages, including all recommended ones, also explicitly
854 
allow commercial use in this way.  A few packages are restricted to
 854 
allow commercial use in this way.  A few packages are restricted to
855 
"non-commercial use"; you should contact the author to clarify whether
 855 
"non-commercial use"; you should contact the author to clarify whether
856 
these may be used or seek the advice of your legal counsel.
 856 
these may be used or seek the advice of your legal counsel.
857 
 
 857 
 
858 
   <p>None of the discussion in this section constitutes legal advice.  The R
 858 
   <p>None of the discussion in this section constitutes legal advice.  The R
859 
Core Team does not provide legal advice under any circumstances.
 859 
Core Team does not provide legal advice under any circumstances.
860 
 
 860 
 
861 
<div class="node">
 861 
<div class="node">
862 
<p><hr>
 862 
<p><hr>
863 
Node:&nbsp;<a name="Why%20is%20R%20named%20R%3f">Why is R named R?</a>,
 863 
Node:&nbsp;<a name="Why%20is%20R%20named%20R%3f">Why is R named R?</a>,
864 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
 864 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Can%20I%20use%20R%20for%20commercial%20purposes%3f">Can I use R for commercial purposes?</a>,
865 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
 865 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Basics">R Basics</a>
866 
<br>
 866 
<br>
867 
</div>
 867 
</div>
868 
 
 868 
 
869 
<h3 class="section">2.12 Why is R named R?</h3>
 869 
<h3 class="section">2.12 Why is R named R?</h3>
870 
 
 870 
 
871 
<p>The name is partly based on the (first) names of the first two R authors
 871 
<p>The name is partly based on the (first) names of the first two R authors
872 
(Robert Gentleman and Ross Ihaka), and partly a play on the name of the
 872 
(Robert Gentleman and Ross Ihaka), and partly a play on the name of the
873 
Bell Labs language `S' (see <a href="#What%20is%20S%3f">What is S?</a>).
 873 
Bell Labs language `S' (see <a href="#What%20is%20S%3f">What is S?</a>).
874 
 
 874 
 
875 
<div class="node">
 875 
<div class="node">
876 
<p><hr>
 876 
<p><hr>
877 
Node:&nbsp;<a name="R%20and%20S">R and S</a>,
 877 
Node:&nbsp;<a name="R%20and%20S">R and S</a>,
878 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Web%20Interfaces">R Web Interfaces</a>,
 878 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Web%20Interfaces">R Web Interfaces</a>,
879 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Basics">R Basics</a>,
 879 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Basics">R Basics</a>,
880 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 880 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
881 
<br>
 881 
<br>
882 
</div>
 882 
</div>
883 
 
 883 
 
884 
<h2 class="chapter">3 R and S</h2>
 884 
<h2 class="chapter">3 R and S</h2>
885 
 
 885 
 
886 
<ul class="menu">
 886 
<ul class="menu">
887 
<li><a accesskey="1" href="#What%20is%20S%3f">What is S?</a>:
887 
<li><a accesskey="1" href="#What%20is%20S%3f">What is S?</a>:
888 
<li><a accesskey="2" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>:
888 
<li><a accesskey="2" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>:
889 
<li><a accesskey="3" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>:
889 
<li><a accesskey="3" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>:
890 
<li><a accesskey="4" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>:
890 
<li><a accesskey="4" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>:
891 
<li><a accesskey="5" href="#What%20is%20R-plus%3f">What is R-plus?</a>:
891 
<li><a accesskey="5" href="#What%20is%20R-plus%3f">What is R-plus?</a>:
892 
</ul>
 892 
</ul>
893 
 
 893 
 
894 
<div class="node">
 894 
<div class="node">
895 
<p><hr>
 895 
<p><hr>
896 
Node:&nbsp;<a name="What%20is%20S%3f">What is S?</a>,
 896 
Node:&nbsp;<a name="What%20is%20S%3f">What is S?</a>,
897 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
 897 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
898 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20S">R and S</a>,
 898 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20S">R and S</a>,
899 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
 899 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
900 
<br>
 900 
<br>
901 
</div>
 901 
</div>
902 
 
 902 
 
903 
<h3 class="section">3.1 What is S?</h3>
 903 
<h3 class="section">3.1 What is S?</h3>
904 
 
 904 
 
905 
<p>S is a very high level language and an environment for data analysis and
 905 
<p>S is a very high level language and an environment for data analysis and
906 
graphics.  In 1998, the Association for Computing Machinery
 906 
graphics.  In 1998, the Association for Computing Machinery
907 
(<small>ACM</small>) presented its Software System Award to John M. Chambers,
 907 
(<small>ACM</small>) presented its Software System Award to John M. Chambers,
908 
the principal designer of S, for
 908 
the principal designer of S, for
909 
 
 909 
 
910 
   <blockquote>
 910 
   <blockquote>
911 
the S system, which has forever altered the way people analyze,
 911 
the S system, which has forever altered the way people analyze,
912 
visualize, and manipulate data <small class="dots">...</small>
 912 
visualize, and manipulate data <small class="dots">...</small>
913 
 
 913 
 
914 
        <p>S is an elegant, widely accepted, and enduring software system, with
 914 
        <p>S is an elegant, widely accepted, and enduring software system, with
915 
conceptual integrity, thanks to the insight, taste, and effort of John
 915 
conceptual integrity, thanks to the insight, taste, and effort of John
916 
Chambers.
916 
Chambers.
917 
</blockquote>
 917 
</blockquote>
918 
 
 918 
 
919 
   <p>The evolution of the S language is characterized by four books by John
 919 
   <p>The evolution of the S language is characterized by four books by John
920 
Chambers and coauthors, which are also the primary references for S.
 920 
Chambers and coauthors, which are also the primary references for S.
921 
 
 921 
 
922 
     <ul>
 922 
     <ul>
923 
<li>Richard A. Becker and John M. Chambers (1984), "S.  An Interactive
 923 
<li>Richard A. Becker and John M. Chambers (1984), "S.  An Interactive
924 
Environment for Data Analysis and Graphics," Monterey: Wadsworth and
 924 
Environment for Data Analysis and Graphics," Monterey: Wadsworth and
925 
Brooks/Cole.
 925 
Brooks/Cole.
926 
 
 926 
 
927 
     <p>This is also referred to as the "<em>Brown Book</em>", and of historical
 927 
     <p>This is also referred to as the "<em>Brown Book</em>", and of historical
928 
interest only.
 928 
interest only.
929 
 
 929 
 
930 
     </p><li>Richard A. Becker, John M. Chambers and Allan R. Wilks (1988), "The New
 930 
     </p><li>Richard A. Becker, John M. Chambers and Allan R. Wilks (1988), "The New
931 
S Language," London: Chapman &amp; Hall.
 931 
S Language," London: Chapman &amp; Hall.
932 
 
 932 
 
933 
     <p>This book is often called the "<em>Blue Book</em>", and introduced what
 933 
     <p>This book is often called the "<em>Blue Book</em>", and introduced what
934 
is now known as S version 2.
 934 
is now known as S version 2.
935 
 
 935 
 
936 
     </p><li>John M. Chambers and Trevor J. Hastie (1992), "Statistical Models in
 936 
     </p><li>John M. Chambers and Trevor J. Hastie (1992), "Statistical Models in
937 
S,"  London: Chapman &amp; Hall.
 937 
S,"  London: Chapman &amp; Hall.
938 
 
 938 
 
939 
     <p>This is also called the "<em>White Book</em>", and introduced S version
 939 
     <p>This is also called the "<em>White Book</em>", and introduced S version
940 
3, which added structures to facilitate statistical modeling in S.
 940 
3, which added structures to facilitate statistical modeling in S.
941 
 
 941 
 
942 
     </p><li>John M. Chambers (1998), "Programming with Data," New York: Springer,
 942 
     </p><li>John M. Chambers (1998), "Programming with Data," New York: Springer,
943 
ISBN 0-387-98503-4
 943 
ISBN 0-387-98503-4
944 
(&lt;<code>http://cm.bell-labs.com/cm/ms/departments/sia/Sbook/</code>&gt;).
 944 
(&lt;<code>http://cm.bell-labs.com/cm/ms/departments/sia/Sbook/</code>&gt;).
945 
 
 945 
 
946 
     <p>This "<em>Green Book</em>" describes version 4 of S, a major revision of
 946 
     <p>This "<em>Green Book</em>" describes version 4 of S, a major revision of
947 
S designed by John Chambers to improve its usefulness at every stage of
 947 
S designed by John Chambers to improve its usefulness at every stage of
948 
the programming process.
948 
the programming process.
949 
</ul>
 949 
</ul>
950 
 
 950 
 
951 
   <p>See <a href="http://cm.bell-labs.com/cm/ms/departments/sia/S/history.html">http://cm.bell-labs.com/cm/ms/departments/sia/S/history.html</a>
 951 
   <p>See <a href="http://cm.bell-labs.com/cm/ms/departments/sia/S/history.html">http://cm.bell-labs.com/cm/ms/departments/sia/S/history.html</a>
952 
for further information on "Stages in the Evolution of S".
 952 
for further information on "Stages in the Evolution of S".
953 
 
 953 
 
954 
   <p>There is a huge amount of user-contributed code for S, available at the
 954 
   <p>There is a huge amount of user-contributed code for S, available at the
955 
<a href="http://lib.stat.cmu.edu/S/">S Repository</a> at <small>CMU</small>.
 955 
<a href="http://lib.stat.cmu.edu/S/">S Repository</a> at <small>CMU</small>.
956 
 
 956 
 
957 
<div class="node">
 957 
<div class="node">
958 
<p><hr>
 958 
<p><hr>
959 
Node:&nbsp;<a name="What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
 959 
Node:&nbsp;<a name="What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
960 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
 960 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
961 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20S%3f">What is S?</a>,
 961 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20S%3f">What is S?</a>,
962 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
 962 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
963 
<br>
 963 
<br>
964 
</div>
 964 
</div>
965 
 
 965 
 
966 
<h3 class="section">3.2 What is <small>S-PLUS</small>?</h3>
 966 
<h3 class="section">3.2 What is <small>S-PLUS</small>?</h3>
967 
 
 967 
 
968 
<p><small>S-PLUS</small> is a value-added version of S sold by Insightful Corporation.
968 
<p><small>S-PLUS</small> is a value-added version of S sold by Insightful Corporation.
969 
Based on the S language, <small>S-PLUS</small> provides functionality in a wide
 969 
Based on the S language, <small>S-PLUS</small> provides functionality in a wide
970 
variety of areas, including robust regression, modern non-parametric
 970 
variety of areas, including robust regression, modern non-parametric
971 
regression, time series, survival analysis, multivariate analysis,
 971 
regression, time series, survival analysis, multivariate analysis,
972 
classical statistical tests, quality control, and graphics drivers.
972 
classical statistical tests, quality control, and graphics drivers.
973 
Add-on modules add additional capabilities for wavelet analysis, spatial
 973 
Add-on modules add additional capabilities for wavelet analysis, spatial
974 
statistics, GARCH models, and design of experiments.
 974 
statistics, GARCH models, and design of experiments.
975 
 
 975 
 
976 
   <p>See the <a href="http://www.insightful.com/products/splus/">Insightful <small>S-PLUS</small> page</a> for further information.
 976 
   <p>See the <a href="http://www.insightful.com/products/splus/">Insightful <small>S-PLUS</small> page</a> for further information.
977 
 
 977 
 
978 
<div class="node">
 978 
<div class="node">
979 
<p><hr>
 979 
<p><hr>
980 
Node:&nbsp;<a name="What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
 980 
Node:&nbsp;<a name="What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
981 
Next:&nbsp;<a rel="next" accesskey="n" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
 981 
Next:&nbsp;<a rel="next" accesskey="n" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
982 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
 982 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
983 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
 983 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
984 
<br>
 984 
<br>
985 
</div>
 985 
</div>
986 
 
 986 
 
987 
<h3 class="section">3.3 What are the differences between R and S?</h3>
 987 
<h3 class="section">3.3 What are the differences between R and S?</h3>
988 
 
 988 
 
989 
<p>We can regard S as a language with three current implementations or
 989 
<p>We can regard S as a language with three current implementations or
990 
"engines", the "old S engine" (S version 3; <small>S-PLUS</small> 3.x and 4.x),
 990 
"engines", the "old S engine" (S version 3; <small>S-PLUS</small> 3.x and 4.x),
991 
the "new S engine" (S version 4; <small>S-PLUS</small> 5.x and above), and R.
991 
the "new S engine" (S version 4; <small>S-PLUS</small> 5.x and above), and R.
992 
Given this understanding, asking for "the differences between R and S"
 992 
Given this understanding, asking for "the differences between R and S"
993 
really amounts to asking for the specifics of the R implementation of
 993 
really amounts to asking for the specifics of the R implementation of
994 
the S language, i.e., the difference between the R and S <em>engines</em>.
 994 
the S language, i.e., the difference between the R and S <em>engines</em>.
995 
 
 995 
 
996 
   <p>For the remainder of this section, "S" refers to the S engines and not
 996 
   <p>For the remainder of this section, "S" refers to the S engines and not
997 
the S language.
 997 
the S language.
998 
 
 998 
 
999 
<ul class="menu">
 999 
<ul class="menu">
1000 
<li><a accesskey="1" href="#Lexical%20scoping">Lexical scoping</a>:
1000 
<li><a accesskey="1" href="#Lexical%20scoping">Lexical scoping</a>:
1001 
<li><a accesskey="2" href="#Models">Models</a>:
1001 
<li><a accesskey="2" href="#Models">Models</a>:
1002 
<li><a accesskey="3" href="#Others">Others</a>:
1002 
<li><a accesskey="3" href="#Others">Others</a>:
1003 
</ul>
 1003 
</ul>
1004 
 
 1004 
 
1005 
<div class="node">
 1005 
<div class="node">
1006 
<p><hr>
 1006 
<p><hr>
1007 
Node:&nbsp;<a name="Lexical%20scoping">Lexical scoping</a>,
 1007 
Node:&nbsp;<a name="Lexical%20scoping">Lexical scoping</a>,
1008 
Next:&nbsp;<a rel="next" accesskey="n" href="#Models">Models</a>,
 1008 
Next:&nbsp;<a rel="next" accesskey="n" href="#Models">Models</a>,
1009 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
 1009 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
1010 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
 1010 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
1011 
<br>
 1011 
<br>
1012 
</div>
 1012 
</div>
1013 
 
 1013 
 
1014 
<h3 class="subsection">3.3.1 Lexical scoping</h4>
 1014 
<h3 class="subsection">3.3.1 Lexical scoping</h4>
1015 
 
 1015 
 
1016 
<p>Contrary to other implementations of the S language, R has adopted the
 1016 
<p>Contrary to other implementations of the S language, R has adopted the
1017 
evaluation model of Scheme.
 1017 
evaluation model of Scheme.
1018 
 
 1018 
 
1019 
   <p>This difference becomes manifest when <em>free</em> variables occur in a
 1019 
   <p>This difference becomes manifest when <em>free</em> variables occur in a
1020 
function.  Free variables are those which are neither formal parameters
 1020 
function.  Free variables are those which are neither formal parameters
1021 
(occurring in the argument list of the function) nor local variables
 1021 
(occurring in the argument list of the function) nor local variables
1022 
(created by assigning to them in the body of the function).  Whereas S
 1022 
(created by assigning to them in the body of the function).  Whereas S
1023 
(like C) by default uses <em>static</em> scoping, R (like Scheme) has
 1023 
(like C) by default uses <em>static</em> scoping, R (like Scheme) has
1024 
adopted <em>lexical</em> scoping.  This means the values of free variables
 1024 
adopted <em>lexical</em> scoping.  This means the values of free variables
1025 
are determined by a set of global variables in S, but in R by the
 1025 
are determined by a set of global variables in S, but in R by the
1026 
bindings that were in effect at the time the function was created.
 1026 
bindings that were in effect at the time the function was created.
1027 
 
 1027 
 
1028 
   <p>Consider the following function:
 1028 
   <p>Consider the following function:
1029 
 
 1029 
 
1030 
<pre class="example">     cube &lt;- function(n) {
 1030 
<pre class="example">     cube &lt;- function(n) {
1031 
       sq &lt;- function() n * n
 1031 
       sq &lt;- function() n * n
1032 
       n * sq()
 1032 
       n * sq()
1033 
     }
 1033 
     }
1034 
     </pre>
 1034 
     </pre>
1035 
 
 1035 
 
1036 
   <p>Under S, <code>sq()</code> does not "know" about the variable <code>n</code>
 1036 
   <p>Under S, <code>sq()</code> does not "know" about the variable <code>n</code>
1037 
unless it is defined globally:
 1037 
unless it is defined globally:
1038 
 
 1038 
 
1039 
<pre class="example">     S&gt; cube(2)
 1039 
<pre class="example">     S&gt; cube(2)
1040 
     Error in sq():  Object "n" not found
 1040 
     Error in sq():  Object "n" not found
1041 
     Dumped
 1041 
     Dumped
1042 
     S&gt; n &lt;- 3
 1042 
     S&gt; n &lt;- 3
1043 
     S&gt; cube(2)
 1043 
     S&gt; cube(2)
1044 
     [1] 18
 1044 
     [1] 18
1045 
     </pre>
 1045 
     </pre>
1046 
 
 1046 
 
1047 
   <p>In R, the "environment" created when <code>cube()</code> was invoked is
 1047 
   <p>In R, the "environment" created when <code>cube()</code> was invoked is
1048 
also looked in:
 1048 
also looked in:
1049 
 
 1049 
 
1050 
<pre class="example">     R&gt; cube(2)
 1050 
<pre class="example">     R&gt; cube(2)
1051 
     [1] 8
 1051 
     [1] 8
1052 
     </pre>
 1052 
     </pre>
1053 
 
 1053 
 
1054 
   <p>As a more "interesting" real-world problem, suppose you want to write
 1054 
   <p>As a more "interesting" real-world problem, suppose you want to write
1055 
a function which returns the density function of the r-th order
 1055 
a function which returns the density function of the r-th order
1056 
statistic from a sample of size n from a (continuous)
 1056 
statistic from a sample of size n from a (continuous)
1057 
distribution.  For simplicity, we shall use both the cdf and pdf of the
 1057 
distribution.  For simplicity, we shall use both the cdf and pdf of the
1058 
distribution as explicit arguments.  (Example compiled from various
 1058 
distribution as explicit arguments.  (Example compiled from various
1059 
postings by Luke Tierney.)
 1059 
postings by Luke Tierney.)
1060 
 
 1060 
 
1061 
   <p>The <small>S-PLUS</small> documentation for <code>call()</code> basically suggests the
 1061 
   <p>The <small>S-PLUS</small> documentation for <code>call()</code> basically suggests the
1062 
following:
 1062 
following:
1063 
 
 1063 
 
1064 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
 1064 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
1065 
       f &lt;- function(x) NULL
 1065 
       f &lt;- function(x) NULL
1066 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
 1066 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
1067 
       PF &lt;- call(substitute(pfun), as.name("x"))
 1067 
       PF &lt;- call(substitute(pfun), as.name("x"))
1068 
       DF &lt;- call(substitute(dfun), as.name("x"))
 1068 
       DF &lt;- call(substitute(dfun), as.name("x"))
1069 
       f[[length(f)]] &lt;-
 1069 
       f[[length(f)]] &lt;-
1070 
         call("*", con,
 1070 
         call("*", con,
1071 
              call("*", call("^", PF, r - 1),
 1071 
              call("*", call("^", PF, r - 1),
1072 
                   call("*", call("^", call("-", 1, PF), n - r),
 1072 
                   call("*", call("^", call("-", 1, PF), n - r),
1073 
                        DF)))
 1073 
                        DF)))
1074 
       f
 1074 
       f
1075 
     }
 1075 
     }
1076 
     </pre>
 1076 
     </pre>
1077 
 
 1077 
 
1078 
<p>Rather tricky, isn't it?  The code uses the fact that in S,
 1078 
<p>Rather tricky, isn't it?  The code uses the fact that in S,
1079 
functions are just lists of special mode with the function body as the
 1079 
functions are just lists of special mode with the function body as the
1080 
last argument, and hence does not work in R (one could make the idea
 1080 
last argument, and hence does not work in R (one could make the idea
1081 
work, though).
 1081 
work, though).
1082 
 
 1082 
 
1083 
   <p>A version which makes heavy use of <code>substitute()</code> and seems to work
 1083 
   <p>A version which makes heavy use of <code>substitute()</code> and seems to work
1084 
under both S and R is
 1084 
under both S and R is
1085 
 
 1085 
 
1086 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
 1086 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
1087 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
 1087 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
1088 
       eval(substitute(function(x) K * PF(x)^a * (1 - PF(x))^b * DF(x),
 1088 
       eval(substitute(function(x) K * PF(x)^a * (1 - PF(x))^b * DF(x),
1089 
                       list(PF = substitute(pfun), DF = substitute(dfun),
 1089 
                       list(PF = substitute(pfun), DF = substitute(dfun),
1090 
                            a = r - 1, b = n - r, K = con)))
 1090 
                            a = r - 1, b = n - r, K = con)))
1091 
     }
 1091 
     }
1092 
     </pre>
 1092 
     </pre>
1093 
 
 1093 
 
1094 
<p>(the <code>eval()</code> is not needed in S).
 1094 
<p>(the <code>eval()</code> is not needed in S).
1095 
 
 1095 
 
1096 
   <p>However, in R there is a much easier solution:
 1096 
   <p>However, in R there is a much easier solution:
1097 
 
 1097 
 
1098 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
 1098 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
1099 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
 1099 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
1100 
       function(x) {
 1100 
       function(x) {
1101 
         con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
 1101 
         con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
1102 
       }
 1102 
       }
1103 
     }
 1103 
     }
1104 
     </pre>
 1104 
     </pre>
1105 
 
 1105 
 
1106 
<p>This seems to be the "natural" implementation, and it works because
 1106 
<p>This seems to be the "natural" implementation, and it works because
1107 
the free variables in the returned function can be looked up in the
 1107 
the free variables in the returned function can be looked up in the
1108 
defining environment (this is lexical scope).
 1108 
defining environment (this is lexical scope).
1109 
 
 1109 
 
1110 
   <p>Note that what you really need is the function <em>closure</em>, i.e., the
 1110 
   <p>Note that what you really need is the function <em>closure</em>, i.e., the
1111 
body along with all variable bindings needed for evaluating it.  Since
 1111 
body along with all variable bindings needed for evaluating it.  Since
1112 
in the above version, the free variables in the value function are not
 1112 
in the above version, the free variables in the value function are not
1113 
modified, you can actually use it in S as well if you abstract out the
 1113 
modified, you can actually use it in S as well if you abstract out the
1114 
closure operation into a function <code>MC()</code> (for "make closure"):
 1114 
closure operation into a function <code>MC()</code> (for "make closure"):
1115 
 
 1115 
 
1116 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
 1116 
<pre class="example">     dorder &lt;- function(n, r, pfun, dfun) {
1117 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
 1117 
       con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
1118 
       MC(function(x) {
 1118 
       MC(function(x) {
1119 
            con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
 1119 
            con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
1120 
          },
 1120 
          },
1121 
          list(con = con, pfun = pfun, dfun = dfun, r = r, n = n))
 1121 
          list(con = con, pfun = pfun, dfun = dfun, r = r, n = n))
1122 
     }
 1122 
     }
1123 
     </pre>
 1123 
     </pre>
1124 
 
 1124 
 
1125 
   <p>Given the appropriate definitions of the closure operator, this works in
 1125 
   <p>Given the appropriate definitions of the closure operator, this works in
1126 
both R and S, and is much "cleaner" than a substitute/eval solution
 1126 
both R and S, and is much "cleaner" than a substitute/eval solution
1127 
(or one which overrules the default scoping rules by using explicit
 1127 
(or one which overrules the default scoping rules by using explicit
1128 
access to evaluation frames, as is of course possible in both R and S).
 1128 
access to evaluation frames, as is of course possible in both R and S).
1129 
 
 1129 
 
1130 
   <p>For R, <code>MC()</code> simply is
 1130 
   <p>For R, <code>MC()</code> simply is
1131 
 
 1131 
 
1132 
<pre class="example">     MC &lt;- function(f, env) f
 1132 
<pre class="example">     MC &lt;- function(f, env) f
1133 
     </pre>
 1133 
     </pre>
1134 
 
 1134 
 
1135 
<p>(lexical scope!), a version for S is
 1135 
<p>(lexical scope!), a version for S is
1136 
 
 1136 
 
1137 
<pre class="example">     MC &lt;- function(f, env = NULL) {
 1137 
<pre class="example">     MC &lt;- function(f, env = NULL) {
1138 
       env &lt;- as.list(env)
 1138 
       env &lt;- as.list(env)
1139 
       if (mode(f) != "function")
 1139 
       if (mode(f) != "function")
1140 
         stop(paste("not a function:", f))
 1140 
         stop(paste("not a function:", f))
1141 
       if (length(env) &gt; 0 &amp;&amp; any(names(env) == ""))
 1141 
       if (length(env) &gt; 0 &amp;&amp; any(names(env) == ""))
1142 
         stop(paste("not all arguments are named:", env))
 1142 
         stop(paste("not all arguments are named:", env))
1143 
       fargs &lt;- if(length(f) &gt; 1) f[1:(length(f) - 1)] else NULL
 1143 
       fargs &lt;- if(length(f) &gt; 1) f[1:(length(f) - 1)] else NULL
1144 
       fargs &lt;- c(fargs, env)
 1144 
       fargs &lt;- c(fargs, env)
1145 
       if (any(duplicated(names(fargs))))
 1145 
       if (any(duplicated(names(fargs))))
1146 
         stop(paste("duplicated arguments:", paste(names(fargs)),
 1146 
         stop(paste("duplicated arguments:", paste(names(fargs)),
1147 
              collapse = ", "))
 1147 
              collapse = ", "))
1148 
       fbody &lt;- f[length(f)]
 1148 
       fbody &lt;- f[length(f)]
1149 
       cf &lt;- c(fargs, fbody)
 1149 
       cf &lt;- c(fargs, fbody)
1150 
       mode(cf) &lt;- "function"
 1150 
       mode(cf) &lt;- "function"
1151 
       return(cf)
 1151 
       return(cf)
1152 
     }
 1152 
     }
1153 
     </pre>
 1153 
     </pre>
1154 
 
 1154 
 
1155 
   <p>Similarly, most optimization (or zero-finding) routines need some
 1155 
   <p>Similarly, most optimization (or zero-finding) routines need some
1156 
arguments to be optimized over and have other parameters that depend on
 1156 
arguments to be optimized over and have other parameters that depend on
1157 
the data but are fixed with respect to optimization.  With R scoping
 1157 
the data but are fixed with respect to optimization.  With R scoping
1158 
rules, this is a trivial problem; simply make up the function with the
 1158 
rules, this is a trivial problem; simply make up the function with the
1159 
required definitions in the same environment and scoping takes care of
 1159 
required definitions in the same environment and scoping takes care of
1160 
it.  With S, one solution is to add an extra parameter to the function
 1160 
it.  With S, one solution is to add an extra parameter to the function
1161 
and to the optimizer to pass in these extras, which however can only
 1161 
and to the optimizer to pass in these extras, which however can only
1162 
work if the optimizer supports this.
 1162 
work if the optimizer supports this.
1163 
 
 1163 
 
1164 
   <p>Lexical scoping allows using function closures and maintaining local
 1164 
   <p>Lexical scoping allows using function closures and maintaining local
1165 
state.  A simple example (taken from Abelson and Sussman) is obtained by
 1165 
state.  A simple example (taken from Abelson and Sussman) is obtained by
1166 
typing <kbd>demo("scoping")</kbd> at the R prompt.  Further information is
 1166 
typing <kbd>demo("scoping")</kbd> at the R prompt.  Further information is
1167 
provided in the standard R reference "R: A Language for Data Analysis
 1167 
provided in the standard R reference "R: A Language for Data Analysis
1168 
and Graphics" (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>) and in Robert
 1168 
and Graphics" (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>) and in Robert
1169 
Gentleman and Ross Ihaka (2000), "Lexical Scope and Statistical
 1169 
Gentleman and Ross Ihaka (2000), "Lexical Scope and Statistical
1170 
Computing", <a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>9</strong>,
 1170 
Computing", <a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>9</strong>,
1171 
491-508.
 1171 
491-508.
1172 
 
 1172 
 
1173 
   <p>Lexical scoping also implies a further major difference.  Whereas S
 1173 
   <p>Lexical scoping also implies a further major difference.  Whereas S
1174 
stores all objects as separate files in a directory somewhere (usually
 1174 
stores all objects as separate files in a directory somewhere (usually
1175 
<code>.Data</code> under the current directory), R does not.  All objects in R
 1175 
<code>.Data</code> under the current directory), R does not.  All objects in R
1176 
are stored internally.  When R is started up it grabs a piece of memory
 1176 
are stored internally.  When R is started up it grabs a piece of memory
1177 
and uses it to store the objects.  R performs its own memory management
 1177 
and uses it to store the objects.  R performs its own memory management
1178 
of this piece of memory, growing and shrinking its size as needed.
1178 
of this piece of memory, growing and shrinking its size as needed.
1179 
Having everything in memory is necessary because it is not really
 1179 
Having everything in memory is necessary because it is not really
1180 
possible to externally maintain all relevant "environments" of
 1180 
possible to externally maintain all relevant "environments" of
1181 
symbol/value pairs.  This difference also seems to make R <em>faster</em>
 1181 
symbol/value pairs.  This difference also seems to make R <em>faster</em>
1182 
than S.
 1182 
than S.
1183 
 
 1183 
 
1184 
   <p>The down side is that if R crashes you will lose all the work for the
 1184 
   <p>The down side is that if R crashes you will lose all the work for the
1185 
current session.  Saving and restoring the memory "images" (the
 1185 
current session.  Saving and restoring the memory "images" (the
1186 
functions and data stored in R's internal memory at any time) can be a
 1186 
functions and data stored in R's internal memory at any time) can be a
1187 
bit slow, especially if they are big.  In S this does not happen,
 1187 
bit slow, especially if they are big.  In S this does not happen,
1188 
because everything is saved in disk files and if you crash nothing is
 1188 
because everything is saved in disk files and if you crash nothing is
1189 
likely to happen to them.  (In fact, one might conjecture that the S
 1189 
likely to happen to them.  (In fact, one might conjecture that the S
1190 
developers felt that the price of changing their approach to persistent
 1190 
developers felt that the price of changing their approach to persistent
1191 
storage just to accommodate lexical scope was far too expensive.)
1191 
storage just to accommodate lexical scope was far too expensive.)
1192 
Hence, when doing important work, you might consider saving often (see
 1192 
Hence, when doing important work, you might consider saving often (see
1193 
<a href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>) to safeguard against possible
 1193 
<a href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>) to safeguard against possible
1194 
crashes.  Other possibilities are logging your sessions, or have your R
 1194 
crashes.  Other possibilities are logging your sessions, or have your R
1195 
commands stored in text files which can be read in using
 1195 
commands stored in text files which can be read in using
1196 
<code>source()</code>.
 1196 
<code>source()</code>.
1197 
 
 1197 
 
1198 
   <blockquote>
 1198 
   <blockquote>
1199 
<strong>Note:</strong>  If you run R from within Emacs (see <a href="#R%20and%20Emacs">R and Emacs</a>),
 1199 
<strong>Note:</strong>  If you run R from within Emacs (see <a href="#R%20and%20Emacs">R and Emacs</a>),
1200 
you can save the contents of the interaction buffer to a file and
 1200 
you can save the contents of the interaction buffer to a file and
1201 
conveniently manipulate it using <code>ess-transcript-mode</code>, as well as
 1201 
conveniently manipulate it using <code>ess-transcript-mode</code>, as well as
1202 
save source copies of all functions and data used.
1202 
save source copies of all functions and data used.
1203 
</blockquote>
 1203 
</blockquote>
1204 
 
 1204 
 
1205 
<div class="node">
 1205 
<div class="node">
1206 
<p><hr>
 1206 
<p><hr>
1207 
Node:&nbsp;<a name="Models">Models</a>,
 1207 
Node:&nbsp;<a name="Models">Models</a>,
1208 
Next:&nbsp;<a rel="next" accesskey="n" href="#Others">Others</a>,
 1208 
Next:&nbsp;<a rel="next" accesskey="n" href="#Others">Others</a>,
1209 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Lexical%20scoping">Lexical scoping</a>,
 1209 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Lexical%20scoping">Lexical scoping</a>,
1210 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
 1210 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
1211 
<br>
 1211 
<br>
1212 
</div>
 1212 
</div>
1213 
 
 1213 
 
1214 
<h3 class="subsection">3.3.2 Models</h4>
 1214 
<h3 class="subsection">3.3.2 Models</h4>
1215 
 
 1215 
 
1216 
<p>There are some differences in the modeling code, such as
 1216 
<p>There are some differences in the modeling code, such as
1217 
 
 1217 
 
1218 
     <ul>
 1218 
     <ul>
1219 
<li>Whereas in S, you would use <code>lm(y ~ x^3)</code> to regress <code>y</code> on
 1219 
<li>Whereas in S, you would use <code>lm(y ~ x^3)</code> to regress <code>y</code> on
1220 
<code>x^3</code>, in R, you have to insulate powers of numeric vectors (using
 1220 
<code>x^3</code>, in R, you have to insulate powers of numeric vectors (using
1221 
<code>I()</code>), i.e., you have to use <code>lm(y ~ I(x^3))</code>.
1221 
<code>I()</code>), i.e., you have to use <code>lm(y ~ I(x^3))</code>.
1222 
<li>The glm family objects are implemented differently in R and S.  The same
 1222 
<li>The glm family objects are implemented differently in R and S.  The same
1223 
functionality is available but the components have different names.
1223 
functionality is available but the components have different names.
1224 
<li>Option <code>na.action</code> is set to <code>"na.omit"</code> by default in R,
 1224 
<li>Option <code>na.action</code> is set to <code>"na.omit"</code> by default in R,
1225 
but not set in S.
1225 
but not set in S.
1226 
<li>Terms objects are stored differently.  In S a terms object is an
 1226 
<li>Terms objects are stored differently.  In S a terms object is an
1227 
expression with attributes, in R it is a formula with attributes.  The
 1227 
expression with attributes, in R it is a formula with attributes.  The
1228 
attributes have the same names but are mostly stored differently.  The
 1228 
attributes have the same names but are mostly stored differently.  The
1229 
major difference in functionality is that a terms object is
 1229 
major difference in functionality is that a terms object is
1230 
subscriptable in S but not in R.  If you can't imagine why this would
 1230 
subscriptable in S but not in R.  If you can't imagine why this would
1231 
matter then you don't need to know.
1231 
matter then you don't need to know.
1232 
<li>Finally, in R <code>y~x+0</code> is an alternative to <code>y~x-1</code> for
 1232 
<li>Finally, in R <code>y~x+0</code> is an alternative to <code>y~x-1</code> for
1233 
specifying a model with no intercept.  Models with no parameters at all
 1233 
specifying a model with no intercept.  Models with no parameters at all
1234 
can be specified by <code>y~0</code>.
1234 
can be specified by <code>y~0</code>.
1235 
</ul>
 1235 
</ul>
1236 
 
 1236 
 
1237 
<div class="node">
 1237 
<div class="node">
1238 
<p><hr>
 1238 
<p><hr>
1239 
Node:&nbsp;<a name="Others">Others</a>,
 1239 
Node:&nbsp;<a name="Others">Others</a>,
1240 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Models">Models</a>,
 1240 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Models">Models</a>,
1241 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
 1241 
Up:&nbsp;<a rel="up" accesskey="u" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
1242 
<br>
 1242 
<br>
1243 
</div>
 1243 
</div>
1244 
 
 1244 
 
1245 
<h3 class="subsection">3.3.3 Others</h4>
 1245 
<h3 class="subsection">3.3.3 Others</h4>
1246 
 
 1246 
 
1247 
<p>Apart from lexical scoping and its implications, R follows the S
 1247 
<p>Apart from lexical scoping and its implications, R follows the S
1248 
language definition in the Blue and White Books as much as possible, and
 1248 
language definition in the Blue and White Books as much as possible, and
1249 
hence really is an "implementation" of S.  There are some intentional
 1249 
hence really is an "implementation" of S.  There are some intentional
1250 
differences where the behavior of S is considered "not clean".  In
 1250 
differences where the behavior of S is considered "not clean".  In
1251 
general, the rationale is that R should help you detect programming
 1251 
general, the rationale is that R should help you detect programming
1252 
errors, while at the same time being as compatible as possible with S.
 1252 
errors, while at the same time being as compatible as possible with S.
1253 
 
 1253 
 
1254 
   <p>Some known differences are the following.
 1254 
   <p>Some known differences are the following.
1255 
 
 1255 
 
1256 
     <ul>
 1256 
     <ul>
1257 
 
 1257 
 
1258 
     <li>In R, if <code>x</code> is a list, then <code>x[i] &lt;- NULL</code> and <code>x[[i]]
 1258 
     <li>In R, if <code>x</code> is a list, then <code>x[i] &lt;- NULL</code> and <code>x[[i]]
1259 
&lt;- NULL</code> remove the specified elements from <code>x</code>.  The first of
 1259 
&lt;- NULL</code> remove the specified elements from <code>x</code>.  The first of
1260 
these is incompatible with S, where it is a no-op.  (Note that you can
 1260 
these is incompatible with S, where it is a no-op.  (Note that you can
1261 
set elements to <code>NULL</code> using <code>x[i] &lt;- list(NULL)</code>.)
 1261 
set elements to <code>NULL</code> using <code>x[i] &lt;- list(NULL)</code>.)
1262 
 
 1262 
 
1263 
     <li>In S, the functions named <code>.First</code> and <code>.Last</code> in the
 1263 
     <li>In S, the functions named <code>.First</code> and <code>.Last</code> in the
1264 
<code>.Data</code> directory can be used for customizing, as they are executed
 1264 
<code>.Data</code> directory can be used for customizing, as they are executed
1265 
at the very beginning and end of a session, respectively.
 1265 
at the very beginning and end of a session, respectively.
1266 
 
 1266 
 
1267 
     <p>In R, the startup mechanism is as follows.  R first sources the system
 1267 
     <p>In R, the startup mechanism is as follows.  R first sources the system
1268 
startup file <code>$R_HOME/library/base/R/Rprofile</code>.  Then, it
 1268 
startup file <code>$R_HOME/library/base/R/Rprofile</code>.  Then, it
1269 
searches for a site-wide startup profile unless the command line option
 1269 
searches for a site-wide startup profile unless the command line option
1270 
<code>--no-site-file</code> was given.  The name of this file is taken from
 1270 
<code>--no-site-file</code> was given.  The name of this file is taken from
1271 
the value of the <code>R_PROFILE</code> environment variable.  If that variable
 1271 
the value of the <code>R_PROFILE</code> environment variable.  If that variable
1272 
is unset, the default is <code>$R_HOME/etc/Rprofile.site</code>
 1272 
is unset, the default is <code>$R_HOME/etc/Rprofile.site</code>
1273 
(<code>$R_HOME/etc/Rprofile</code> in versions prior to 1.4.0).  This
 1273 
(<code>$R_HOME/etc/Rprofile</code> in versions prior to 1.4.0).  This
1274 
code is loaded in package <strong>base</strong>.  Then, unless
 1274 
code is loaded in package <strong>base</strong>.  Then, unless
1275 
<code>--no-init-file</code> was given, R searches for a file called
 1275 
<code>--no-init-file</code> was given, R searches for a file called
1276 
<code>.Rprofile</code> in the current directory or in the user's home
 1276 
<code>.Rprofile</code> in the current directory or in the user's home
1277 
directory (in that order) and sources it into the user workspace.  It
 1277 
directory (in that order) and sources it into the user workspace.  It
1278 
then loads a saved image of the user workspace from <code>.RData</code> in
 1278 
then loads a saved image of the user workspace from <code>.RData</code> in
1279 
case there is one (unless <code>--no-restore</code> was specified).  If
 1279 
case there is one (unless <code>--no-restore</code> was specified).  If
1280 
needed, the functions <code>.First()</code> and <code>.Last()</code> should be
 1280 
needed, the functions <code>.First()</code> and <code>.Last()</code> should be
1281 
defined in the appropriate startup profiles.
 1281 
defined in the appropriate startup profiles.
1282 
 
 1282 
 
1283 
     </p><li>In R, <code>T</code> and <code>F</code> are just variables being set to <code>TRUE</code>
 1283 
     </p><li>In R, <code>T</code> and <code>F</code> are just variables being set to <code>TRUE</code>
1284 
and <code>FALSE</code>, respectively, but are not reserved words as in S and
 1284 
and <code>FALSE</code>, respectively, but are not reserved words as in S and
1285 
hence can be overwritten by the user.  (This helps e.g. when you have
 1285 
hence can be overwritten by the user.  (This helps e.g. when you have
1286 
factors with levels <code>"T"</code> or <code>"F"</code>.)  Hence, when writing code
 1286 
factors with levels <code>"T"</code> or <code>"F"</code>.)  Hence, when writing code
1287 
you should always use <code>TRUE</code> and <code>FALSE</code>.
 1287 
you should always use <code>TRUE</code> and <code>FALSE</code>.
1288 
 
 1288 
 
1289 
     <li>In R, <code>dyn.load()</code> can only load <em>shared objects</em>, as created
 1289 
     <li>In R, <code>dyn.load()</code> can only load <em>shared objects</em>, as created
1290 
for example by <kbd>R CMD SHLIB</kbd>.
 1290 
for example by <kbd>R CMD SHLIB</kbd>.
1291 
 
 1291 
 
1292 
     <li>In R, <code>attach()</code> currently only works for lists and data frames,
 1292 
     <li>In R, <code>attach()</code> currently only works for lists and data frames,
1293 
but not for directories.  (In fact, <code>attach()</code> also works for R
 1293 
but not for directories.  (In fact, <code>attach()</code> also works for R
1294 
data files created with <code>save()</code>, which is analogous to attaching
 1294 
data files created with <code>save()</code>, which is analogous to attaching
1295 
directories in S.)  Also, you cannot attach at position 1.
 1295 
directories in S.)  Also, you cannot attach at position 1.
1296 
 
 1296 
 
1297 
     <li>Categories do not exist in R, and never will as they are deprecated now
 1297 
     <li>Categories do not exist in R, and never will as they are deprecated now
1298 
in S.  Use factors instead.
 1298 
in S.  Use factors instead.
1299 
 
 1299 
 
1300 
     <li>In R, <code>For()</code> loops are not necessary and hence not supported.
 1300 
     <li>In R, <code>For()</code> loops are not necessary and hence not supported.
1301 
 
 1301 
 
1302 
     <li>In R, <code>assign()</code> uses the argument <code>envir=</code> rather than
 1302 
     <li>In R, <code>assign()</code> uses the argument <code>envir=</code> rather than
1303 
<code>where=</code> as in S.
 1303 
<code>where=</code> as in S.
1304 
 
 1304 
 
1305 
     <li>The random number generators are different, and the seeds have different
 1305 
     <li>The random number generators are different, and the seeds have different
1306 
length.
 1306 
length.
1307 
 
 1307 
 
1308 
     <li>R passes integer objects to C as <code>int *</code> rather than <code>long *</code>
 1308 
     <li>R passes integer objects to C as <code>int *</code> rather than <code>long *</code>
1309 
as in S.
 1309 
as in S.
1310 
 
 1310 
 
1311 
     <li>R has no single precision storage mode.  However, as of version 0.65.1,
 1311 
     <li>R has no single precision storage mode.  However, as of version 0.65.1,
1312 
there is a single precision interface to C/FORTRAN subroutines.
 1312 
there is a single precision interface to C/FORTRAN subroutines.
1313 
 
 1313 
 
1314 
     <li>By default, <code>ls()</code> returns the names of the objects in the current
 1314 
     <li>By default, <code>ls()</code> returns the names of the objects in the current
1315 
(under R) and global (under S) environment, respectively.  For example,
 1315 
(under R) and global (under S) environment, respectively.  For example,
1316 
given
 1316 
given
1317 
 
 1317 
 
1318 
     <pre class="example">          x &lt;- 1; fun &lt;- function() {y &lt;- 1; ls()}
 1318 
     <pre class="example">          x &lt;- 1; fun &lt;- function() {y &lt;- 1; ls()}
1319 
          </pre>
 1319 
          </pre>
1320 
 
 1320 
 
1321 
     <p>then <code>fun()</code> returns <code>"y"</code> in R and <code>"x"</code> (together with
 1321 
     <p>then <code>fun()</code> returns <code>"y"</code> in R and <code>"x"</code> (together with
1322 
the rest of the global environment) in S.
 1322 
the rest of the global environment) in S.
1323 
 
 1323 
 
1324 
     </p><li>R allows for zero-extent matrices (and arrays, i.e., some elements of
 1324 
     </p><li>R allows for zero-extent matrices (and arrays, i.e., some elements of
1325 
the <code>dim</code> attribute vector can be 0).  This has been determined a
 1325 
the <code>dim</code> attribute vector can be 0).  This has been determined a
1326 
useful feature as it helps reducing the need for special-case tests for
 1326 
useful feature as it helps reducing the need for special-case tests for
1327 
empty subsets.  For example, if <code>x</code> is a matrix, <code>x[, FALSE]</code>
 1327 
empty subsets.  For example, if <code>x</code> is a matrix, <code>x[, FALSE]</code>
1328 
is not <code>NULL</code> but a "matrix" with 0 columns.  Hence, such objects
 1328 
is not <code>NULL</code> but a "matrix" with 0 columns.  Hence, such objects
1329 
need to be tested for by checking whether their <code>length()</code> is zero
 1329 
need to be tested for by checking whether their <code>length()</code> is zero
1330 
(which works in both R and S), and not using <code>is.null()</code>.
 1330 
(which works in both R and S), and not using <code>is.null()</code>.
1331 
 
 1331 
 
1332 
     <li>Named vectors are considered vectors in R but not in S (e.g.,
 1332 
     <li>Named vectors are considered vectors in R but not in S (e.g.,
1333 
<code>is.vector(c(a = 1:3))</code> returns <code>FALSE</code> in S and <code>TRUE</code>
 1333 
<code>is.vector(c(a = 1:3))</code> returns <code>FALSE</code> in S and <code>TRUE</code>
1334 
in R).
 1334 
in R).
1335 
 
 1335 
 
1336 
     <li>Data frames are not considered as matrices in R (i.e., if <code>DF</code> is a
 1336 
     <li>Data frames are not considered as matrices in R (i.e., if <code>DF</code> is a
1337 
data frame, then <code>is.matrix(DF)</code> returns <code>FALSE</code> in R and
 1337 
data frame, then <code>is.matrix(DF)</code> returns <code>FALSE</code> in R and
1338 
<code>TRUE</code> in S).
 1338 
<code>TRUE</code> in S).
1339 
 
 1339 
 
1340 
     <li>R by default uses treatment contrasts in the unordered case, whereas S
 1340 
     <li>R by default uses treatment contrasts in the unordered case, whereas S
1341 
uses the Helmert ones.  This is a deliberate difference reflecting the
 1341 
uses the Helmert ones.  This is a deliberate difference reflecting the
1342 
opinion that treatment contrasts are more natural.
 1342 
opinion that treatment contrasts are more natural.
1343 
 
 1343 
 
1344 
     <li>In R, the argument of a replacement function which corresponds to the
 1344 
     <li>In R, the argument of a replacement function which corresponds to the
1345 
right hand side must be named <code>value</code>.  E.g., <code>f(a) &lt;- b</code> is
 1345 
right hand side must be named <code>value</code>.  E.g., <code>f(a) &lt;- b</code> is
1346 
evaluated as <code>a &lt;- "f&lt;-"(a, value = b)</code>.  S always takes the last
 1346 
evaluated as <code>a &lt;- "f&lt;-"(a, value = b)</code>.  S always takes the last
1347 
argument, irrespective of its name.
 1347 
argument, irrespective of its name.
1348 
 
 1348 
 
1349 
     <li>In S, <code>substitute()</code> searches for names for substitution in the
 1349 
     <li>In S, <code>substitute()</code> searches for names for substitution in the
1350 
given expression in three places: the actual and the default arguments
 1350 
given expression in three places: the actual and the default arguments
1351 
of the matching call, and the local frame (in that order).  R looks in
 1351 
of the matching call, and the local frame (in that order).  R looks in
1352 
the local frame only, with the special rule to use a "promise" if a
 1352 
the local frame only, with the special rule to use a "promise" if a
1353 
variable is not evaluated.  Since the local frame is initialized with
 1353 
variable is not evaluated.  Since the local frame is initialized with
1354 
the actual arguments or the default expressions, this is usually
 1354 
the actual arguments or the default expressions, this is usually
1355 
equivalent to S, until assignment takes place.
 1355 
equivalent to S, until assignment takes place.
1356 
 
 1356 
 
1357 
     <li>In S, the index variable in a <code>for()</code> loop is local to the inside
 1357 
     <li>In S, the index variable in a <code>for()</code> loop is local to the inside
1358 
of the loop.  In R it is local to the environment where the <code>for()</code>
 1358 
of the loop.  In R it is local to the environment where the <code>for()</code>
1359 
statement is executed.
 1359 
statement is executed.
1360 
 
 1360 
 
1361 
     <li>In S, <code>tapply(simplify=TRUE)</code> returns a vector where R returns a
 1361 
     <li>In S, <code>tapply(simplify=TRUE)</code> returns a vector where R returns a
1362 
one-dimensional array (which can have named dimnames).
 1362 
one-dimensional array (which can have named dimnames).
1363 
 
 1363 
 
1364 
     <li>In S(-<small>PLUS</small>) the C locale is used, whereas in R the current
 1364 
     <li>In S(-<small>PLUS</small>) the C locale is used, whereas in R the current
1365 
operating system locale is used for determining which characters are
 1365 
operating system locale is used for determining which characters are
1366 
alphanumeric and how they are sorted.  This affects the set of valid
 1366 
alphanumeric and how they are sorted.  This affects the set of valid
1367 
names for R objects (for example accented chars may be allowed in R) and
 1367 
names for R objects (for example accented chars may be allowed in R) and
1368 
ordering in sorts and comparisons (such as whether <code>"aA" &lt; "Bb"</code> is
 1368 
ordering in sorts and comparisons (such as whether <code>"aA" &lt; "Bb"</code> is
1369 
true or false).  From version 1.2.0 the locale can be (re-)set in R by
 1369 
true or false).  From version 1.2.0 the locale can be (re-)set in R by
1370 
the <code>Sys.setlocale()</code> function.
 1370 
the <code>Sys.setlocale()</code> function.
1371 
 
 1371 
 
1372 
     <li>In S, <code>missing(</code><var>arg</var><code>)</code> remains <code>TRUE</code> if <var>arg</var> is
 1372 
     <li>In S, <code>missing(</code><var>arg</var><code>)</code> remains <code>TRUE</code> if <var>arg</var> is
1373 
subsequently modified; in R it doesn't.
 1373 
subsequently modified; in R it doesn't.
1374 
 
 1374 
 
1375 
     <li>From R version 1.3.0, <code>data.frame</code> strips <code>I()</code> when creating
 1375 
     <li>From R version 1.3.0, <code>data.frame</code> strips <code>I()</code> when creating
1376 
(column) names.
 1376 
(column) names.
1377 
 
 1377 
 
1378 
     <li>In R, the string <code>"NA"</code> is not treated as a missing value in a
 1378 
     <li>In R, the string <code>"NA"</code> is not treated as a missing value in a
1379 
character variable.  Use <code>as.character(NA)</code> to create a missing
 1379 
character variable.  Use <code>as.character(NA)</code> to create a missing
1380 
character value.
 1380 
character value.
1381 
 
 1381 
 
1382 
     <li>R disallows repeated formal arguments in function calls.
 1382 
     <li>R disallows repeated formal arguments in function calls.
1383 
 
 1383 
 
1384 
   </ul>
 1384 
   </ul>
1385 
 
 1385 
 
1386 
   <p>There are also differences which are not intentional, and result from
 1386 
   <p>There are also differences which are not intentional, and result from
1387 
missing or incorrect code in R.  The developers would appreciate hearing
 1387 
missing or incorrect code in R.  The developers would appreciate hearing
1388 
about any deficiencies you may find (in a written report fully
 1388 
about any deficiencies you may find (in a written report fully
1389 
documenting the difference as you see it).  Of course, it would be
 1389 
documenting the difference as you see it).  Of course, it would be
1390 
useful if you were to implement the change yourself and make sure it
 1390 
useful if you were to implement the change yourself and make sure it
1391 
works.
 1391 
works.
1392 
 
 1392 
 
1393 
<div class="node">
 1393 
<div class="node">
1394 
<p><hr>
 1394 
<p><hr>
1395 
Node:&nbsp;<a name="Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
 1395 
Node:&nbsp;<a name="Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
1396 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20R-plus%3f">What is R-plus?</a>,
 1396 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20is%20R-plus%3f">What is R-plus?</a>,
1397 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
 1397 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
1398 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
 1398 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
1399 
<br>
 1399 
<br>
1400 
</div>
 1400 
</div>
1401 
 
 1401 
 
1402 
<h3 class="section">3.4 Is there anything R can do that <small>S-PLUS</small> cannot?</h3>
 1402 
<h3 class="section">3.4 Is there anything R can do that <small>S-PLUS</small> cannot?</h3>
1403 
 
 1403 
 
1404 
<p>Since almost anything you can do in R has source code that you could
 1404 
<p>Since almost anything you can do in R has source code that you could
1405 
port to <small>S-PLUS</small> with little effort there will never be much you can do
 1405 
port to <small>S-PLUS</small> with little effort there will never be much you can do
1406 
in R that you couldn't do in <small>S-PLUS</small> if you wanted to.  (Note that
 1406 
in R that you couldn't do in <small>S-PLUS</small> if you wanted to.  (Note that
1407 
using lexical scoping may simplify matters considerably, though.)
 1407 
using lexical scoping may simplify matters considerably, though.)
1408 
 
 1408 
 
1409 
   <p>R offers several graphics features that <small>S-PLUS</small> does not, such as finer
 1409 
   <p>R offers several graphics features that <small>S-PLUS</small> does not, such as finer
1410 
handling of line types, more convenient color handling (via palettes),
 1410 
handling of line types, more convenient color handling (via palettes),
1411 
gamma correction for color, and, most importantly, mathematical
 1411 
gamma correction for color, and, most importantly, mathematical
1412 
annotation in plot texts, via input expressions reminiscent of TeX
 1412 
annotation in plot texts, via input expressions reminiscent of TeX
1413 
constructs.  See the help page for <code>plotmath</code>, which features an
 1413 
constructs.  See the help page for <code>plotmath</code>, which features an
1414 
impressive on-line example.  More details can be found in Paul Murrell
 1414 
impressive on-line example.  More details can be found in Paul Murrell
1415 
and Ross Ihaka (2000), "An Approach to Providing Mathematical
 1415 
and Ross Ihaka (2000), "An Approach to Providing Mathematical
1416 
Annotation in Plots", <a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>9</strong>,
 1416 
Annotation in Plots", <a href="http://www.amstat.org/publications/jcgs/"><em>Journal of Computational and Graphical Statistics</em></a>, <strong>9</strong>,
1417 
582-599.
 1417 
582-599.
1418 
 
 1418 
 
1419 
<div class="node">
 1419 
<div class="node">
1420 
<p><hr>
 1420 
<p><hr>
1421 
Node:&nbsp;<a name="What%20is%20R-plus%3f">What is R-plus?</a>,
 1421 
Node:&nbsp;<a name="What%20is%20R-plus%3f">What is R-plus?</a>,
1422 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
 1422 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
1423 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
 1423 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20S">R and S</a>
1424 
<br>
 1424 
<br>
1425 
</div>
 1425 
</div>
1426 
 
 1426 
 
1427 
<h3 class="section">3.5 What is R-plus?</h3>
 1427 
<h3 class="section">3.5 What is R-plus?</h3>
1428 
 
 1428 
 
1429 
<p>There is no such thing.
 1429 
<p>There is no such thing.
1430 
 
 1430 
 
1431 
<div class="node">
 1431 
<div class="node">
1432 
<p><hr>
 1432 
<p><hr>
1433 
Node:&nbsp;<a name="R%20Web%20Interfaces">R Web Interfaces</a>,
 1433 
Node:&nbsp;<a name="R%20Web%20Interfaces">R Web Interfaces</a>,
1434 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
 1434 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
1435 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20S">R and S</a>,
 1435 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20S">R and S</a>,
1436 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 1436 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
1437 
<br>
 1437 
<br>
1438 
</div>
 1438 
</div>
1439 
 
 1439 
 
1440 
<h2 class="chapter">4 R Web Interfaces</h2>
 1440 
<h2 class="chapter">4 R Web Interfaces</h2>
1441 
 
 1441 
 
1442 
<p><strong>Rweb</strong> is developed and maintained by
 1442 
<p><strong>Rweb</strong> is developed and maintained by
1443 
<a href="mailto:jeff@math.montana.edu">Jeff Banfield</a>.  The
 1443 
<a href="mailto:jeff@math.montana.edu">Jeff Banfield</a>.  The
1444 
<a href="http://www.math.montana.edu/Rweb/">Rweb Home Page</a> provides access
 1444 
<a href="http://www.math.montana.edu/Rweb/">Rweb Home Page</a> provides access
1445 
to all three versions of Rweb--a simple text entry form that returns
 1445 
to all three versions of Rweb--a simple text entry form that returns
1446 
output and graphs, a more sophisticated Javascript version that provides
 1446 
output and graphs, a more sophisticated Javascript version that provides
1447 
a multiple window environment, and a set of point and click modules that
 1447 
a multiple window environment, and a set of point and click modules that
1448 
are useful for introductory statistics courses and require no knowledge
 1448 
are useful for introductory statistics courses and require no knowledge
1449 
of the R language.  All of the Rweb versions can analyze Web accessible
 1449 
of the R language.  All of the Rweb versions can analyze Web accessible
1450 
datasets if a <small>URL</small> is provided.
 1450 
datasets if a <small>URL</small> is provided.
1451 
 
 1451 
 
1452 
   <p>The paper "Rweb: Web-based Statistical Analysis", providing a detailed
 1452 
   <p>The paper "Rweb: Web-based Statistical Analysis", providing a detailed
1453 
explanation of the different versions of Rweb and an overview of how
 1453 
explanation of the different versions of Rweb and an overview of how
1454 
Rweb works, was published in the Journal of Statistical Software
 1454 
Rweb works, was published in the Journal of Statistical Software
1455 
(<a href="http://www.stat.ucla.edu/journals/jss/v04/i01/">http://www.stat.ucla.edu/journals/jss/v04/i01/</a>).
 1455 
(<a href="http://www.stat.ucla.edu/journals/jss/v04/i01/">http://www.stat.ucla.edu/journals/jss/v04/i01/</a>).
1456 
 
 1456 
 
1457 
   <a href="mailto:ulfi@cs.tu-berlin.de">Ulf Bartel</a> is working on
 1457 
   <a href="mailto:ulfi@cs.tu-berlin.de">Ulf Bartel</a> is working on
1458 
<strong>R-Online</strong>, a simple on-line programming environment for R which
 1458 
<strong>R-Online</strong>, a simple on-line programming environment for R which
1459 
intends to make the first steps in statistical programming with R
 1459 
intends to make the first steps in statistical programming with R
1460 
(especially with time series) as easy as possible.  There is no need for
 1460 
(especially with time series) as easy as possible.  There is no need for
1461 
a local installation since the only requirement for the user is a
 1461 
a local installation since the only requirement for the user is a
1462 
JavaScript capable browser.  See <a href="http://osvisions.com/r-online/">http://osvisions.com/r-online/</a>
 1462 
JavaScript capable browser.  See <a href="http://osvisions.com/r-online/">http://osvisions.com/r-online/</a>
1463 
for more information.
 1463 
for more information.
1464 
 
 1464 
 
1465 
   <a href="mailto:http://www.warwick.ac.uk/go/dfirth">David Firth</a> has written
 1465 
   <a href="mailto:http://www.warwick.ac.uk/go/dfirth">David Firth</a> has written
1466 
<strong>CGIwithR</strong>, an R add-on package available from <small>CRAN</small>.  It
 1466 
<strong>CGIwithR</strong>, an R add-on package available from <small>CRAN</small>.  It
1467 
provides some simple extensions to R to facilitate running R scripts
 1467 
provides some simple extensions to R to facilitate running R scripts
1468 
through the CGI interface to a web server.  It is easily installed using
 1468 
through the CGI interface to a web server.  It is easily installed using
1469 
Apache under Linux and in principle should run on any platform that
 1469 
Apache under Linux and in principle should run on any platform that
1470 
supports R and a web server provided that the installer has the
 1470 
supports R and a web server provided that the installer has the
1471 
necessary security permissions.
 1471 
necessary security permissions.
1472 
 
 1472 
 
1473 
   <p><strong>Rcgi</strong> is a CGI WWW interface to R by <a href="mailto:mjr@dsl.pipex.com">MJ Ray</a>.  It had the ability to use "embedded code": you could mix
 1473 
   <p><strong>Rcgi</strong> is a CGI WWW interface to R by <a href="mailto:mjr@dsl.pipex.com">MJ Ray</a>.  It had the ability to use "embedded code": you could mix
1474 
user input and code, allowing the <small>HTML</small> author to do anything from
 1474 
user input and code, allowing the <small>HTML</small> author to do anything from
1475 
load in data sets to enter most of the commands for users without
 1475 
load in data sets to enter most of the commands for users without
1476 
writing CGI scripts.  Graphical output was possible in PostScript or GIF
 1476 
writing CGI scripts.  Graphical output was possible in PostScript or GIF
1477 
formats and the executed code was presented to the user for revision.
1477 
formats and the executed code was presented to the user for revision.
1478 
However, it is not clear if the project is still active.
1478 
However, it is not clear if the project is still active.
1479 
Currently, a modified version of <strong>Rcgi</strong> by
 1479 
Currently, a modified version of <strong>Rcgi</strong> by
1480 
<a href="mailto:mai@ms.uky.edu">Mai Zhou</a> (actually, two versions: one with
 1480 
<a href="mailto:mai@ms.uky.edu">Mai Zhou</a> (actually, two versions: one with
1481 
(bitmap) graphics and one without) as well as the original code are
 1481 
(bitmap) graphics and one without) as well as the original code are
1482 
available from <a href="http://www.ms.uky.edu/~statweb">http://www.ms.uky.edu/~statweb</a>.
 1482 
available from <a href="http://www.ms.uky.edu/~statweb">http://www.ms.uky.edu/~statweb</a>.
1483 
 
 1483 
 
1484 
<div class="node">
 1484 
<div class="node">
1485 
<p><hr>
 1485 
<p><hr>
1486 
Node:&nbsp;<a name="R%20Add-On%20Packages">R Add-On Packages</a>,
 1486 
Node:&nbsp;<a name="R%20Add-On%20Packages">R Add-On Packages</a>,
1487 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20and%20Emacs">R and Emacs</a>,
 1487 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20and%20Emacs">R and Emacs</a>,
1488 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Web%20Interfaces">R Web Interfaces</a>,
 1488 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Web%20Interfaces">R Web Interfaces</a>,
1489 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 1489 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
1490 
<br>
 1490 
<br>
1491 
</div>
 1491 
</div>
1492 
 
 1492 
 
1493 
<h2 class="chapter">5 R Add-On Packages</h2>
 1493 
<h2 class="chapter">5 R Add-On Packages</h2>
1494 
 
 1494 
 
1495 
<ul class="menu">
 1495 
<ul class="menu">
1496 
<li><a accesskey="1" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>:
1496 
<li><a accesskey="1" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>:
1497 
<li><a accesskey="2" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>:
1497 
<li><a accesskey="2" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>:
1498 
<li><a accesskey="3" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>:
1498 
<li><a accesskey="3" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>:
1499 
<li><a accesskey="4" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>:
1499 
<li><a accesskey="4" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>:
1500 
<li><a accesskey="5" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>:
1500 
<li><a accesskey="5" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>:
1501 
<li><a accesskey="6" href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>:
1501 
<li><a accesskey="6" href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>:
1502 
</ul>
 1502 
</ul>
1503 
 
 1503 
 
1504 
<div class="node">
 1504 
<div class="node">
1505 
<p><hr>
 1505 
<p><hr>
1506 
Node:&nbsp;<a name="Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
 1506 
Node:&nbsp;<a name="Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
1507 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
 1507 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
1508 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
 1508 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
1509 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 1509 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
1510 
<br>
 1510 
<br>
1511 
</div>
 1511 
</div>
1512 
 
 1512 
 
1513 
<h3 class="section">5.1 Which add-on packages exist for R?</h3>
 1513 
<h3 class="section">5.1 Which add-on packages exist for R?</h3>
1514 
 
 1514 
 
1515 
<ul class="menu">
 1515 
<ul class="menu">
1516 
<li><a accesskey="1" href="#Add-on%20packages%20in%20R">Add-on packages in R</a>:
1516 
<li><a accesskey="1" href="#Add-on%20packages%20in%20R">Add-on packages in R</a>:
1517 
<li><a accesskey="2" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>:
1517 
<li><a accesskey="2" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>:
1518 
<li><a accesskey="3" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>:
1518 
<li><a accesskey="3" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>:
1519 
<li><a accesskey="4" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>:
1519 
<li><a accesskey="4" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>:
1520 
<li><a accesskey="5" href="#Other%20add-on%20packages">Other add-on packages</a>:
1520 
<li><a accesskey="5" href="#Other%20add-on%20packages">Other add-on packages</a>:
1521 
</ul>
 1521 
</ul>
1522 
 
 1522 
 
1523 
<div class="node">
 1523 
<div class="node">
1524 
<p><hr>
 1524 
<p><hr>
1525 
Node:&nbsp;<a name="Add-on%20packages%20in%20R">Add-on packages in R</a>,
 1525 
Node:&nbsp;<a name="Add-on%20packages%20in%20R">Add-on packages in R</a>,
1526 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
 1526 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
1527 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
 1527 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
1528 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
 1528 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
1529 
<br>
 1529 
<br>
1530 
</div>
 1530 
</div>
1531 
 
 1531 
 
1532 
<h3 class="subsection">5.1.1 Add-on packages in R</h4>
 1532 
<h3 class="subsection">5.1.1 Add-on packages in R</h4>
1533 
 
 1533 
 
1534 
<p>The R distribution comes with the following extra packages:
 1534 
<p>The R distribution comes with the following extra packages:
1535 
 
 1535 
 
1536 
     <dl>
 1536 
     <dl>
1537 
<dt><strong>ctest</strong>
 1537 
<dt><strong>ctest</strong>
1538 
     <dd>A collection of Classical TESTs, including the Ansari-Bradley, Bartlett,
 1538 
     <dd>A collection of Classical TESTs, including the Ansari-Bradley, Bartlett,
1539 
chi-squared, Fisher, Kruskal-Wallis, Kolmogorov-Smirnov, t, and
 1539 
chi-squared, Fisher, Kruskal-Wallis, Kolmogorov-Smirnov, t, and
1540 
Wilcoxon tests.
1540 
Wilcoxon tests.
1541 
<br><dt><strong>eda</strong>
 1541 
<br><dt><strong>eda</strong>
1542 
     <dd>Exploratory Data Analysis.  Currently only contains functions for robust
 1542 
     <dd>Exploratory Data Analysis.  Currently only contains functions for robust
1543 
line fitting, and median polish and smoothing.
1543 
line fitting, and median polish and smoothing.
1544 
<br><dt><strong>grid</strong>
 1544 
<br><dt><strong>grid</strong>
1545 
     <dd>A rewrite of the graphics layout capabilities, plus some support for
 1545 
     <dd>A rewrite of the graphics layout capabilities, plus some support for
1546 
interaction.
1546 
interaction.
1547 
(Added in R 1.8.0).
1547 
(Added in R 1.8.0).
1548 
<br><dt><strong>lqs</strong>
 1548 
<br><dt><strong>lqs</strong>
1549 
     <dd>Resistant regression and covariance estimation.
1549 
     <dd>Resistant regression and covariance estimation.
1550 
<br><dt><strong>methods</strong>
 1550 
<br><dt><strong>methods</strong>
1551 
     <dd>Formally defined methods and classes for R objects, plus other
 1551 
     <dd>Formally defined methods and classes for R objects, plus other
1552 
programming tools, as described in the Green Book.
1552 
programming tools, as described in the Green Book.
1553 
<br><dt><strong>mle</strong>
 1553 
<br><dt><strong>mle</strong>
1554 
     <dd>Generic (smooth) likelihood maximization and profiling.
1554 
     <dd>Generic (smooth) likelihood maximization and profiling.
1555 
(Added in R 1.8.0).
1555 
(Added in R 1.8.0).
1556 
<br><dt><strong>modreg</strong>
 1556 
<br><dt><strong>modreg</strong>
1557 
     <dd>MODern REGression: smoothing and local methods.
1557 
     <dd>MODern REGression: smoothing and local methods.
1558 
<br><dt><strong>mva</strong>
 1558 
<br><dt><strong>mva</strong>
1559 
     <dd>MultiVariate Analysis.  Currently contains code for principal
 1559 
     <dd>MultiVariate Analysis.  Currently contains code for principal
1560 
components, canonical correlations, metric multidimensional scaling,
 1560 
components, canonical correlations, metric multidimensional scaling,
1561 
factor analysis, and hierarchical and k-means clustering.
1561 
factor analysis, and hierarchical and k-means clustering.
1562 
<br><dt><strong>nls</strong>
 1562 
<br><dt><strong>nls</strong>
1563 
     <dd>Nonlinear regression routines.
1563 
     <dd>Nonlinear regression routines.
1564 
<br><dt><strong>splines</strong>
 1564 
<br><dt><strong>splines</strong>
1565 
     <dd>Regression spline functions and classes.
1565 
     <dd>Regression spline functions and classes.
1566 
<br><dt><strong>stepfun</strong>
 1566 
<br><dt><strong>stepfun</strong>
1567 
     <dd>Code for dealing with STEP FUNctions, including empirical cumulative
 1567 
     <dd>Code for dealing with STEP FUNctions, including empirical cumulative
1568 
distribution functions.
1568 
distribution functions.
1569 
<br><dt><strong>tcltk</strong>
 1569 
<br><dt><strong>tcltk</strong>
1570 
     <dd>Interface and language bindings to Tcl/Tk <small>GUI</small> elements.
1570 
     <dd>Interface and language bindings to Tcl/Tk <small>GUI</small> elements.
1571 
<br><dt><strong>tools</strong>
 1571 
<br><dt><strong>tools</strong>
1572 
     <dd>Tools for package development and administration.
1572 
     <dd>Tools for package development and administration.
1573 
<br><dt><strong>ts</strong>
 1573 
<br><dt><strong>ts</strong>
1574 
     <dd>Time Series.
1574 
     <dd>Time Series.
1575 
</dl>
 1575 
</dl>
1576 
   In R 1.9, <strong>base</strong> will be split into the four packages
 1576 
   In R 1.9, <strong>base</strong> will be split into the four packages
1577 
<strong>base</strong>, <strong>graphics</strong>, <strong>stats</strong>, and <strong>utils</strong>.
1577 
<strong>base</strong>, <strong>graphics</strong>, <strong>stats</strong>, and <strong>utils</strong>.
1578 
Packages <strong>ctest</strong>, <strong>eda</strong>, <strong>modreg</strong>, <strong>mva</strong>,
 1578 
Packages <strong>ctest</strong>, <strong>eda</strong>, <strong>modreg</strong>, <strong>mva</strong>,
1579 
<strong>nls</strong>, <strong>stepfun</strong> and <strong>ts</strong> will be merged into
 1579 
<strong>nls</strong>, <strong>stepfun</strong> and <strong>ts</strong> will be merged into
1580 
<strong>stats</strong>, package <strong>lqs</strong> returned to the recommended package
 1580 
<strong>stats</strong>, package <strong>lqs</strong> returned to the recommended package
1581 
<strong>MASS</strong>, and package <strong>mle</strong> moved to <strong>stats4</strong>.
 1581 
<strong>MASS</strong>, and package <strong>mle</strong> moved to <strong>stats4</strong>.
1582 
 
 1582 
 
1583 
<div class="node">
 1583 
<div class="node">
1584 
<p><hr>
 1584 
<p><hr>
1585 
Node:&nbsp;<a name="Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
 1585 
Node:&nbsp;<a name="Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
1586 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
 1586 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
1587 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20in%20R">Add-on packages in R</a>,
 1587 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20in%20R">Add-on packages in R</a>,
1588 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
 1588 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
1589 
<br>
 1589 
<br>
1590 
</div>
 1590 
</div>
1591 
 
 1591 
 
1592 
<h3 class="subsection">5.1.2 Add-on packages from <small>CRAN</small></h4>
 1592 
<h3 class="subsection">5.1.2 Add-on packages from <small>CRAN</small></h4>
1593 
 
 1593 
 
1594 
<p>The following packages are available from the <small>CRAN</small> <code>src/contrib</code>
 1594 
<p>The following packages are available from the <small>CRAN</small> <code>src/contrib</code>
1595 
area.  (Packages denoted as <em>Recommended</em> are to be included in all
 1595 
area.  (Packages denoted as <em>Recommended</em> are to be included in all
1596 
binary distributions of R.)
 1596 
binary distributions of R.)
1597 
 
 1597 
 
1598 
     <dl>
 1598 
     <dl>
1599 
<dt><strong>AlgDesign</strong>
 1599 
<dt><strong>AlgDesign</strong>
1600 
     <dd>Algorithmic experimental designs.  Calculates exact and approximate
 1600 
     <dd>Algorithmic experimental designs.  Calculates exact and approximate
1601 
theory experimental designs for D, A, and I criteria.
1601 
theory experimental designs for D, A, and I criteria.
1602 
<br><dt><strong>AnalyzeFMRI</strong>
 1602 
<br><dt><strong>AnalyzeFMRI</strong>
1603 
     <dd>Functions for I/O, visualisation and analysis of functional Magnetic
 1603 
     <dd>Functions for I/O, visualisation and analysis of functional Magnetic
1604 
Resonance Imaging (fMRI) datasets stored in the ANALYZE format.
1604 
Resonance Imaging (fMRI) datasets stored in the ANALYZE format.
1605 
<br><dt><strong>Bhat</strong>
 1605 
<br><dt><strong>Bhat</strong>
1606 
     <dd>Functions for general likelihood exploration (MLE, MCMC, CIs).
1606 
     <dd>Functions for general likelihood exploration (MLE, MCMC, CIs).
1607 
<br><dt><strong>BradleyTerry</strong>
 1607 
<br><dt><strong>BradleyTerry</strong>
1608 
     <dd>Specify and fit the Bradley-Terry model and structured versions.
1608 
     <dd>Specify and fit the Bradley-Terry model and structured versions.
1609 
<br><dt><strong>BsMD</strong>
 1609 
<br><dt><strong>BsMD</strong>
1610 
     <dd>Bayes screening and model discrimination follow-up designs.
1610 
     <dd>Bayes screening and model discrimination follow-up designs.
1611 
<br><dt><strong>CDNmoney</strong>
 1611 
<br><dt><strong>CDNmoney</strong>
1612 
     <dd>Components of Canadian monetary aggregates.
1612 
     <dd>Components of Canadian monetary aggregates.
1613 
<br><dt><strong>CGIwithR</strong>
 1613 
<br><dt><strong>CGIwithR</strong>
1614 
     <dd>Facilities for the use of R to write CGI scripts.
1614 
     <dd>Facilities for the use of R to write CGI scripts.
1615 
<br><dt><strong>CircStats</strong>
 1615 
<br><dt><strong>CircStats</strong>
1616 
     <dd>Circular Statistics, from "Topics in Circular Statistics" by S. Rao
 1616 
     <dd>Circular Statistics, from "Topics in Circular Statistics" by S. Rao
1617 
Jammalamadaka and A. SenGupta, 2001, World Scientific.
1617 
Jammalamadaka and A. SenGupta, 2001, World Scientific.
1618 
<br><dt><strong>CoCoAn</strong>
 1618 
<br><dt><strong>CoCoAn</strong>
1619 
     <dd>Constrained Correspondence Analysis.
1619 
     <dd>Constrained Correspondence Analysis.
1620 
<br><dt><strong>DAAG</strong>
 1620 
<br><dt><strong>DAAG</strong>
1621 
     <dd>Various data sets used in examples and exercises in "Data Analysis and
 1621 
     <dd>Various data sets used in examples and exercises in "Data Analysis and
1622 
Graphics Using R" by John H. Maindonald and W. John Brown, 2003.
1622 
Graphics Using R" by John H. Maindonald and W. John Brown, 2003.
1623 
<br><dt><strong>DBI</strong>
 1623 
<br><dt><strong>DBI</strong>
1624 
     <dd>A common database interface (DBI) class and method definitions.  All
 1624 
     <dd>A common database interface (DBI) class and method definitions.  All
1625 
classes in this package are virtual and need to be extended by the
 1625 
classes in this package are virtual and need to be extended by the
1626 
various DBMS implementations.
1626 
various DBMS implementations.
1627 
<br><dt><strong>Davies</strong>
 1627 
<br><dt><strong>Davies</strong>
1628 
     <dd>Functions for the Davies quantile function and the Generalized Lambda
 1628 
     <dd>Functions for the Davies quantile function and the Generalized Lambda
1629 
distribution.
1629 
distribution.
1630 
<br><dt><strong>Design</strong>
 1630 
<br><dt><strong>Design</strong>
1631 
     <dd>Regression modeling, testing, estimation, validation, graphics,
 1631 
     <dd>Regression modeling, testing, estimation, validation, graphics,
1632 
prediction, and typesetting by storing enhanced model design attributes
 1632 
prediction, and typesetting by storing enhanced model design attributes
1633 
in the fit.  Design is a collection of about 180 functions that assist
 1633 
in the fit.  Design is a collection of about 180 functions that assist
1634 
and streamline modeling, especially for biostatistical and epidemiologic
 1634 
and streamline modeling, especially for biostatistical and epidemiologic
1635 
applications.  It also contains new functions for binary and ordinal
 1635 
applications.  It also contains new functions for binary and ordinal
1636 
logistic regression models and the Buckley-James multiple regression
 1636 
logistic regression models and the Buckley-James multiple regression
1637 
model for right-censored responses, and implements penalized maximum
 1637 
model for right-censored responses, and implements penalized maximum
1638 
likelihood estimation for logistic and ordinary linear models.  Design
 1638 
likelihood estimation for logistic and ordinary linear models.  Design
1639 
works with almost any regression model, but it was especially written to
 1639 
works with almost any regression model, but it was especially written to
1640 
work with logistic regression, Cox regression, accelerated failure time
 1640 
work with logistic regression, Cox regression, accelerated failure time
1641 
models, ordinary linear models, and the Buckley-James model.
1641 
models, ordinary linear models, and the Buckley-James model.
1642 
<br><dt><strong>Devore5</strong>
 1642 
<br><dt><strong>Devore5</strong>
1643 
     <dd>Data sets and sample analyses from "Probability and Statistics for
 1643 
     <dd>Data sets and sample analyses from "Probability and Statistics for
1644 
Engineering and the Sciences (5th ed)" by Jay L. Devore, 2000, Duxbury.
1644 
Engineering and the Sciences (5th ed)" by Jay L. Devore, 2000, Duxbury.
1645 
<br><dt><strong>Devore6</strong>
 1645 
<br><dt><strong>Devore6</strong>
1646 
     <dd>Data sets and sample analyses from "Probability and Statistics for
 1646 
     <dd>Data sets and sample analyses from "Probability and Statistics for
1647 
Engineering and the Sciences (6th ed)" by Jay L. Devore, 2003, Duxbury.
1647 
Engineering and the Sciences (6th ed)" by Jay L. Devore, 2003, Duxbury.
1648 
<br><dt><strong>EMV</strong>
 1648 
<br><dt><strong>EMV</strong>
1649 
     <dd>Estimation of missing values in a matrix by a k-th nearest
 1649 
     <dd>Estimation of missing values in a matrix by a k-th nearest
1650 
neighboors algorithm.
1650 
neighboors algorithm.
1651 
<br><dt><strong>GRASS</strong>
 1651 
<br><dt><strong>GRASS</strong>
1652 
     <dd>An interface between the GRASS geographical information system and R,
 1652 
     <dd>An interface between the GRASS geographical information system and R,
1653 
based on starting R from within the GRASS environment and chosen
 1653 
based on starting R from within the GRASS environment and chosen
1654 
LOCATION_NAME and MAPSET.  Wrapper and helper functions are provided for
 1654 
LOCATION_NAME and MAPSET.  Wrapper and helper functions are provided for
1655 
a range of R functions to match the interface metadata structures.
1655 
a range of R functions to match the interface metadata structures.
1656 
<br><dt><strong>GenKern</strong>
 1656 
<br><dt><strong>GenKern</strong>
1657 
     <dd>Functions for generating and manipulating generalised binned kernel
 1657 
     <dd>Functions for generating and manipulating generalised binned kernel
1658 
density estimates.
1658 
density estimates.
1659 
<br><dt><strong>GeneTS</strong>
 1659 
<br><dt><strong>GeneTS</strong>
1660 
     <dd>A package for analysing multiple gene expression time series data.
1660 
     <dd>A package for analysing multiple gene expression time series data.
1661 
Currently, implements methods for cell cycle analysis and for inferring
 1661 
Currently, implements methods for cell cycle analysis and for inferring
1662 
large sparse graphical Gaussian models.
1662 
large sparse graphical Gaussian models.
1663 
<br><dt><strong>HI</strong>
 1663 
<br><dt><strong>HI</strong>
1664 
     <dd>Simulation from distributions supported by nested hyperplanes.
1664 
     <dd>Simulation from distributions supported by nested hyperplanes.
1665 
<br><dt><strong>Hmisc</strong>
 1665 
<br><dt><strong>Hmisc</strong>
1666 
     <dd>Functions useful for data analysis, high-level graphics, utility
 1666 
     <dd>Functions useful for data analysis, high-level graphics, utility
1667 
operations, functions for computing sample size and power, importing
 1667 
operations, functions for computing sample size and power, importing
1668 
datasets, imputing missing values, advanced table making, variable
 1668 
datasets, imputing missing values, advanced table making, variable
1669 
clustering, character string manipulation, conversion of S objects to
 1669 
clustering, character string manipulation, conversion of S objects to
1670 
LaTeX code, recoding variables, and bootstrap repeated measures
 1670 
LaTeX code, recoding variables, and bootstrap repeated measures
1671 
analysis.
1671 
analysis.
1672 
<br><dt><strong>HyperbolicDist</strong>
 1672 
<br><dt><strong>HyperbolicDist</strong>
1673 
     <dd>Basic functions for the hyperbolic distribution: probability density
 1673 
     <dd>Basic functions for the hyperbolic distribution: probability density
1674 
function, distribution function, quantile function, a routine for
 1674 
function, distribution function, quantile function, a routine for
1675 
generating observations from the hyperbolic, and a function for fitting
 1675 
generating observations from the hyperbolic, and a function for fitting
1676 
the hyperbolic distribution to data.
1676 
the hyperbolic distribution to data.
1677 
<br><dt><strong>ISwR</strong>
 1677 
<br><dt><strong>ISwR</strong>
1678 
     <dd>Data sets for "Introductory Statistics with R" by Peter Dalgaard,
 1678 
     <dd>Data sets for "Introductory Statistics with R" by Peter Dalgaard,
1679 
2002, Springer.
1679 
2002, Springer.
1680 
<br><dt><strong>KMsurv</strong>
 1680 
<br><dt><strong>KMsurv</strong>
1681 
     <dd>Data sets and functions for "Survival Analysis, Techniques for Censored
 1681 
     <dd>Data sets and functions for "Survival Analysis, Techniques for Censored
1682 
and Truncated Data" by Klein and Moeschberger, 1997, Springer.
1682 
and Truncated Data" by Klein and Moeschberger, 1997, Springer.
1683 
<br><dt><strong>KernSmooth</strong>
 1683 
<br><dt><strong>KernSmooth</strong>
1684 
     <dd>Functions for kernel smoothing (and density estimation) corresponding to
 1684 
     <dd>Functions for kernel smoothing (and density estimation) corresponding to
1685 
the book "Kernel Smoothing" by M. P. Wand and M. C. Jones, 1995.
1685 
the book "Kernel Smoothing" by M. P. Wand and M. C. Jones, 1995.
1686 
<em>Recommended</em>.
1686 
<em>Recommended</em>.
1687 
<br><dt><strong>MASS</strong>
 1687 
<br><dt><strong>MASS</strong>
1688 
     <dd>Functions and datasets from the main package of Venables and Ripley,
 1688 
     <dd>Functions and datasets from the main package of Venables and Ripley,
1689 
"Modern Applied Statistics with S".  Contained in the <code>VR</code>
 1689 
"Modern Applied Statistics with S".  Contained in the <code>VR</code>
1690 
bundle.  <em>Recommended</em>.
1690 
bundle.  <em>Recommended</em>.
1691 
<br><dt><strong>MCMCpack</strong>
 1691 
<br><dt><strong>MCMCpack</strong>
1692 
     <dd>Markov chain Monte Carlo (MCMC) package: functions for posterior
 1692 
     <dd>Markov chain Monte Carlo (MCMC) package: functions for posterior
1693 
simulation for a number of statistical models.
1693 
simulation for a number of statistical models.
1694 
<br><dt><strong>MPV</strong>
 1694 
<br><dt><strong>MPV</strong>
1695 
     <dd>Data sets from the book "Introduction to Linear Regression Analysis"
 1695 
     <dd>Data sets from the book "Introduction to Linear Regression Analysis"
1696 
by D. C. Montgomery, E. A. Peck, and C. G. Vining, 2001, John Wiley and
 1696 
by D. C. Montgomery, E. A. Peck, and C. G. Vining, 2001, John Wiley and
1697 
Sons.
1697 
Sons.
1698 
<br><dt><strong>Matrix</strong>
 1698 
<br><dt><strong>Matrix</strong>
1699 
     <dd>A Matrix package.
1699 
     <dd>A Matrix package.
1700 
<br><dt><strong>NISTnls</strong>
 1700 
<br><dt><strong>NISTnls</strong>
1701 
     <dd>A set of test nonlinear least squares examples from <small>NIST</small>, the
 1701 
     <dd>A set of test nonlinear least squares examples from <small>NIST</small>, the
1702 
U.S. National Institute for Standards and Technology.
1702 
U.S. National Institute for Standards and Technology.
1703 
<br><dt><strong>Oarray</strong>
 1703 
<br><dt><strong>Oarray</strong>
1704 
     <dd>Arrays with arbitrary offsets.
1704 
     <dd>Arrays with arbitrary offsets.
1705 
<br><dt><strong>PHYLOGR</strong>
 1705 
<br><dt><strong>PHYLOGR</strong>
1706 
     <dd>Manipulation and analysis of phylogenetically simulated data sets (as
 1706 
     <dd>Manipulation and analysis of phylogenetically simulated data sets (as
1707 
obtained from PDSIMUL in package PDAP) and phylogenetically-based
 1707 
obtained from PDSIMUL in package PDAP) and phylogenetically-based
1708 
analyses using GLS.
1708 
analyses using GLS.
1709 
<br><dt><strong>PTAk</strong>
 1709 
<br><dt><strong>PTAk</strong>
1710 
     <dd>A multiway method to decompose a tensor (array) of any order, as a
 1710 
     <dd>A multiway method to decompose a tensor (array) of any order, as a
1711 
generalisation of SVD also supporting non-identity metrics and
 1711 
generalisation of SVD also supporting non-identity metrics and
1712 
penalisations.  Also includes some other multiway methods.
1712 
penalisations.  Also includes some other multiway methods.
1713 
<br><dt><strong>R2HTML</strong>
 1713 
<br><dt><strong>R2HTML</strong>
1714 
     <dd>Functions for exporting R objects &amp; graphics in an <small>HTML</small> document.
1714 
     <dd>Functions for exporting R objects &amp; graphics in an <small>HTML</small> document.
1715 
<br><dt><strong>R2WinBUGS</strong>
 1715 
<br><dt><strong>R2WinBUGS</strong>
1716 
     <dd>Running WinBUGS from R: call a BUGS model, summarize inferences and
 1716 
     <dd>Running WinBUGS from R: call a BUGS model, summarize inferences and
1717 
convergence in a table and graph, and save the simulations in arrays for
 1717 
convergence in a table and graph, and save the simulations in arrays for
1718 
easy access in R.
1718 
easy access in R.
1719 
<br><dt><strong>RArcInfo</strong>
 1719 
<br><dt><strong>RArcInfo</strong>
1720 
     <dd>Functions to import Arc/Info V7.x coverages and data.
1720 
     <dd>Functions to import Arc/Info V7.x coverages and data.
1721 
<br><dt><strong>RColorBrewer</strong>
 1721 
<br><dt><strong>RColorBrewer</strong>
1722 
     <dd>ColorBrewer palettes for drawing nice maps shaded according to a
 1722 
     <dd>ColorBrewer palettes for drawing nice maps shaded according to a
1723 
variable.
1723 
variable.
1724 
<br><dt><strong>RMySQL</strong>
 1724 
<br><dt><strong>RMySQL</strong>
1725 
     <dd>An interface between R and the MySQL database system.
1725 
     <dd>An interface between R and the MySQL database system.
1726 
<br><dt><strong>RODBC</strong>
 1726 
<br><dt><strong>RODBC</strong>
1727 
     <dd>An <small>ODBC</small> database interface.
1727 
     <dd>An <small>ODBC</small> database interface.
1728 
<br><dt><strong>ROracle</strong>
 1728 
<br><dt><strong>ROracle</strong>
1729 
     <dd>Oracle Database Interface driver for R.  Uses the ProC/C++ embedded SQL.
1729 
     <dd>Oracle Database Interface driver for R.  Uses the ProC/C++ embedded SQL.
1730 
<br><dt><strong>RQuantLib</strong>
 1730 
<br><dt><strong>RQuantLib</strong>
1731 
     <dd>Provides access to (some) of the QuantLib functions from within R;
 1731 
     <dd>Provides access to (some) of the QuantLib functions from within R;
1732 
currently limited to some Option pricing and analysis functions.  The
 1732 
currently limited to some Option pricing and analysis functions.  The
1733 
QuantLib project aims to provide a comprehensive software framework for
 1733 
QuantLib project aims to provide a comprehensive software framework for
1734 
quantitative finance.
1734 
quantitative finance.
1735 
<br><dt><strong>RSQLite</strong>
 1735 
<br><dt><strong>RSQLite</strong>
1736 
     <dd>Database Interface R driver for SQLite.  Embeds the SQLite database
 1736 
     <dd>Database Interface R driver for SQLite.  Embeds the SQLite database
1737 
engine in R.
1737 
engine in R.
1738 
<br><dt><strong>RSvgDevice</strong>
 1738 
<br><dt><strong>RSvgDevice</strong>
1739 
     <dd>A graphics device for R that uses the new w3.org <small>XML</small> standard for
 1739 
     <dd>A graphics device for R that uses the new w3.org <small>XML</small> standard for
1740 
Scalable Vector Graphics.
1740 
Scalable Vector Graphics.
1741 
<br><dt><strong>RadioSonde</strong>
 1741 
<br><dt><strong>RadioSonde</strong>
1742 
     <dd>A collection of programs for reading and plotting SKEW-T,log p diagrams
 1742 
     <dd>A collection of programs for reading and plotting SKEW-T,log p diagrams
1743 
and wind profiles for data collected by radiosondes (the typical weather
 1743 
and wind profiles for data collected by radiosondes (the typical weather
1744 
balloon-borne instrument).
1744 
balloon-borne instrument).
1745 
<br><dt><strong>RandomFields</strong>
 1745 
<br><dt><strong>RandomFields</strong>
1746 
     <dd>Creating random fields using various methods.
1746 
     <dd>Creating random fields using various methods.
1747 
<br><dt><strong>Rcmdr</strong>
 1747 
<br><dt><strong>Rcmdr</strong>
1748 
     <dd>A platform-independent basic-statistics GUI (graphical user interface)
 1748 
     <dd>A platform-independent basic-statistics GUI (graphical user interface)
1749 
for R, based on the <strong>tcltk</strong> package.
1749 
for R, based on the <strong>tcltk</strong> package.
1750 
<br><dt><strong>RmSQL</strong>
 1750 
<br><dt><strong>RmSQL</strong>
1751 
     <dd>An interface between R and the mSQL database system.
1751 
     <dd>An interface between R and the mSQL database system.
1752 
<br><dt><strong>Rwave</strong>
 1752 
<br><dt><strong>Rwave</strong>
1753 
     <dd>An environment for the time-frequency analysis of 1-D signals (and
 1753 
     <dd>An environment for the time-frequency analysis of 1-D signals (and
1754 
especially for the wavelet and Gabor transforms of noisy signals), based
 1754 
especially for the wavelet and Gabor transforms of noisy signals), based
1755 
on the book "Practical Time-Frequency Analysis: Gabor and Wavelet
 1755 
on the book "Practical Time-Frequency Analysis: Gabor and Wavelet
1756 
Transforms with an Implementation in S" by Rene Carmona, Wen L. Hwang
 1756 
Transforms with an Implementation in S" by Rene Carmona, Wen L. Hwang
1757 
and Bruno Torresani, 1998, Academic Press.
1757 
and Bruno Torresani, 1998, Academic Press.
1758 
<br><dt><strong>SASmixed</strong>
 1758 
<br><dt><strong>SASmixed</strong>
1759 
     <dd>Data sets and sample linear mixed effects analyses corresponding to the
 1759 
     <dd>Data sets and sample linear mixed effects analyses corresponding to the
1760 
examples in "SAS System for Mixed Models" by R. C. Littell,
 1760 
examples in "SAS System for Mixed Models" by R. C. Littell,
1761 
G. A. Milliken, W. W. Stroup and R. D. Wolfinger, 1996, SAS Institute.
1761 
G. A. Milliken, W. W. Stroup and R. D. Wolfinger, 1996, SAS Institute.
1762 
<br><dt><strong>SenSrivastava</strong>
 1762 
<br><dt><strong>SenSrivastava</strong>
1763 
     <dd>Collection of datasets from "Regression Analysis, Theory, Methods and
 1763 
     <dd>Collection of datasets from "Regression Analysis, Theory, Methods and
1764 
Applications" by A. Sen and M. Srivastava, 1990, Springer-Verlag.
1764 
Applications" by A. Sen and M. Srivastava, 1990, Springer-Verlag.
1765 
<br><dt><strong>SoPhy</strong>
 1765 
<br><dt><strong>SoPhy</strong>
1766 
     <dd>Soil Physics Tools: simulation of water flux and solute transport in
 1766 
     <dd>Soil Physics Tools: simulation of water flux and solute transport in
1767 
soil.
1767 
soil.
1768 
<br><dt><strong>SparseM</strong>
 1768 
<br><dt><strong>SparseM</strong>
1769 
     <dd>Basic linear algebra for sparse matrices.
1769 
     <dd>Basic linear algebra for sparse matrices.
1770 
<br><dt><strong>StatDataML</strong>
 1770 
<br><dt><strong>StatDataML</strong>
1771 
     <dd>Read and write StatDataML.
1771 
     <dd>Read and write StatDataML.
1772 
<br><dt><strong>SuppDists</strong>
 1772 
<br><dt><strong>SuppDists</strong>
1773 
     <dd>Ten distributions supplementing those built into R (Inverse Gauss,
 1773 
     <dd>Ten distributions supplementing those built into R (Inverse Gauss,
1774 
Kruskal-Wallis, Kendall's Tau, Friedman's chi squared, Spearman's rho,
 1774 
Kruskal-Wallis, Kendall's Tau, Friedman's chi squared, Spearman's rho,
1775 
maximum F ratio, the Pearson product moment correlation coefficiant,
 1775 
maximum F ratio, the Pearson product moment correlation coefficiant,
1776 
Johnson distributions, normal scores and generalized hypergeometric
 1776 
Johnson distributions, normal scores and generalized hypergeometric
1777 
distributions).
1777 
distributions).
1778 
<br><dt><strong>VLMC</strong>
 1778 
<br><dt><strong>VLMC</strong>
1779 
     <dd>Functions, classes &amp; methods for estimation, prediction, and simulation
 1779 
     <dd>Functions, classes &amp; methods for estimation, prediction, and simulation
1780 
(bootstrap) of VLMC (Variable Length Markov Chain) models.
1780 
(bootstrap) of VLMC (Variable Length Markov Chain) models.
1781 
<br><dt><strong>VaR</strong>
 1781 
<br><dt><strong>VaR</strong>
1782 
     <dd>Methods for calculation of Value at Risk (VaR).
1782 
     <dd>Methods for calculation of Value at Risk (VaR).
1783 
<br><dt><strong>XML</strong>
 1783 
<br><dt><strong>XML</strong>
1784 
     <dd>Facilities for reading <small>XML</small> documents and DTDs.
1784 
     <dd>Facilities for reading <small>XML</small> documents and DTDs.
1785 
<br><dt><strong>abind</strong>
 1785 
<br><dt><strong>abind</strong>
1786 
     <dd>Combine multi-dimensional arrays.
1786 
     <dd>Combine multi-dimensional arrays.
1787 
<br><dt><strong>acepack</strong>
 1787 
<br><dt><strong>acepack</strong>
1788 
     <dd>ACE (Alternating Conditional Expectations) and AVAS (Additivity and
 1788 
     <dd>ACE (Alternating Conditional Expectations) and AVAS (Additivity and
1789 
VAriance Stabilization for regression) methods for selecting regression
 1789 
VAriance Stabilization for regression) methods for selecting regression
1790 
transformations.
1790 
transformations.
1791 
<br><dt><strong>adapt</strong>
 1791 
<br><dt><strong>adapt</strong>
1792 
     <dd>Adaptive quadrature in up to 20 dimensions.
1792 
     <dd>Adaptive quadrature in up to 20 dimensions.
1793 
<br><dt><strong>ade4</strong>
 1793 
<br><dt><strong>ade4</strong>
1794 
     <dd>Multivariate data analysis and graphical display.
1794 
     <dd>Multivariate data analysis and graphical display.
1795 
<br><dt><strong>agce</strong>
 1795 
<br><dt><strong>agce</strong>
1796 
     <dd>Analysis of growth curve experiments.
1796 
     <dd>Analysis of growth curve experiments.
1797 
<br><dt><strong>akima</strong>
 1797 
<br><dt><strong>akima</strong>
1798 
     <dd>Linear or cubic spline interpolation for irregularly gridded data.
1798 
     <dd>Linear or cubic spline interpolation for irregularly gridded data.
1799 
<br><dt><strong>amap</strong>
 1799 
<br><dt><strong>amap</strong>
1800 
     <dd>Another Multidimensional Analysis Package.
1800 
     <dd>Another Multidimensional Analysis Package.
1801 
<br><dt><strong>anm</strong>
 1801 
<br><dt><strong>anm</strong>
1802 
     <dd>Analog model for statistical/empirical downscaling.
1802 
     <dd>Analog model for statistical/empirical downscaling.
1803 
<br><dt><strong>ape</strong>
 1803 
<br><dt><strong>ape</strong>
1804 
     <dd>Analyses of Phylogenetics and Evolution, providing functions for reading
 1804 
     <dd>Analyses of Phylogenetics and Evolution, providing functions for reading
1805 
and plotting phylogenetic trees in parenthetic format (standard Newick
 1805 
and plotting phylogenetic trees in parenthetic format (standard Newick
1806 
format), analyses of comparative data in a phylogenetic framework,
 1806 
format), analyses of comparative data in a phylogenetic framework,
1807 
analyses of diversification and macroevolution, computing distances from
 1807 
analyses of diversification and macroevolution, computing distances from
1808 
allelic and nucleotide data, reading nucleotide sequences from GenBank
 1808 
allelic and nucleotide data, reading nucleotide sequences from GenBank
1809 
via internet, and several tools such as Mantel's test, computation of
 1809 
via internet, and several tools such as Mantel's test, computation of
1810 
minimum spanning tree, or the population parameter theta based on
 1810 
minimum spanning tree, or the population parameter theta based on
1811 
various approaches.
1811 
various approaches.
1812 
<br><dt><strong>ash</strong>
 1812 
<br><dt><strong>ash</strong>
1813 
     <dd>David Scott's ASH routines for 1D and 2D density estimation.
1813 
     <dd>David Scott's ASH routines for 1D and 2D density estimation.
1814 
<br><dt><strong>assist</strong>
 1814 
<br><dt><strong>assist</strong>
1815 
     <dd>A suite of functions implementing smoothing splines.
1815 
     <dd>A suite of functions implementing smoothing splines.
1816 
<br><dt><strong>asypow</strong>
 1816 
<br><dt><strong>asypow</strong>
1817 
     <dd>A set of routines that calculate power and related quantities utilizing
 1817 
     <dd>A set of routines that calculate power and related quantities utilizing
1818 
asymptotic likelihood ratio methods.
1818 
asymptotic likelihood ratio methods.
1819 
<br><dt><strong>aws</strong>
 1819 
<br><dt><strong>aws</strong>
1820 
     <dd>Functions to perform adaptive weights smoothing.
1820 
     <dd>Functions to perform adaptive weights smoothing.
1821 
<br><dt><strong>bim</strong>
 1821 
<br><dt><strong>bim</strong>
1822 
     <dd>Bayesian interval mapping diagnostics:  functions to interpret QTLCart
 1822 
     <dd>Bayesian interval mapping diagnostics:  functions to interpret QTLCart
1823 
and Bmapqtl samples.
1823 
and Bmapqtl samples.
1824 
<br><dt><strong>bindata</strong>
 1824 
<br><dt><strong>bindata</strong>
1825 
     <dd>Generation of correlated artificial binary data.
1825 
     <dd>Generation of correlated artificial binary data.
1826 
<br><dt><strong>blighty</strong>
 1826 
<br><dt><strong>blighty</strong>
1827 
     <dd>Function for drawing the coastline of the United Kingdom.
1827 
     <dd>Function for drawing the coastline of the United Kingdom.
1828 
<br><dt><strong>boolean</strong>
 1828 
<br><dt><strong>boolean</strong>
1829 
     <dd>Boolean logit and probit: a procedure for testing Boolean hypotheses.
1829 
     <dd>Boolean logit and probit: a procedure for testing Boolean hypotheses.
1830 
<br><dt><strong>boot</strong>
 1830 
<br><dt><strong>boot</strong>
1831 
     <dd>Functions and datasets for bootstrapping from the book "Bootstrap
 1831 
     <dd>Functions and datasets for bootstrapping from the book "Bootstrap
1832 
Methods and Their Applications" by A. C. Davison and D. V. Hinkley,
 1832 
Methods and Their Applications" by A. C. Davison and D. V. Hinkley,
1833 
1997, Cambridge University Press.  <em>Recommended</em>.
1833 
1997, Cambridge University Press.  <em>Recommended</em>.
1834 
<br><dt><strong>bootstrap</strong>
 1834 
<br><dt><strong>bootstrap</strong>
1835 
     <dd>Software (bootstrap, cross-validation, jackknife), data and errata for
 1835 
     <dd>Software (bootstrap, cross-validation, jackknife), data and errata for
1836 
the book "An Introduction to the Bootstrap" by B. Efron and
 1836 
the book "An Introduction to the Bootstrap" by B. Efron and
1837 
R. Tibshirani, 1993, Chapman and Hall.
1837 
R. Tibshirani, 1993, Chapman and Hall.
1838 
<br><dt><strong>bqtl</strong>
 1838 
<br><dt><strong>bqtl</strong>
1839 
     <dd>QTL mapping toolkit for inbred crosses and recombinant inbred lines.
1839 
     <dd>QTL mapping toolkit for inbred crosses and recombinant inbred lines.
1840 
Includes maximum likelihood and Bayesian tools.
1840 
Includes maximum likelihood and Bayesian tools.
1841 
<br><dt><strong>brlr</strong>
 1841 
<br><dt><strong>brlr</strong>
1842 
     <dd>Bias-reduced logistic regression: fits logistic regression models by
 1842 
     <dd>Bias-reduced logistic regression: fits logistic regression models by
1843 
maximum penalized likelihood.
1843 
maximum penalized likelihood.
1844 
<br><dt><strong>car</strong>
 1844 
<br><dt><strong>car</strong>
1845 
     <dd>Companion to Applied Regression, containing functions for applied
 1845 
     <dd>Companion to Applied Regression, containing functions for applied
1846 
regession, linear models, and generalized linear models, with an
 1846 
regession, linear models, and generalized linear models, with an
1847 
emphasis on regression diagnostics, particularly graphical diagnostic
 1847 
emphasis on regression diagnostics, particularly graphical diagnostic
1848 
methods.
1848 
methods.
1849 
<br><dt><strong>cat</strong>
 1849 
<br><dt><strong>cat</strong>
1850 
     <dd>Analysis of categorical-variable datasets with missing values.
1850 
     <dd>Analysis of categorical-variable datasets with missing values.
1851 
<br><dt><strong>cclust</strong>
 1851 
<br><dt><strong>cclust</strong>
1852 
     <dd>Convex clustering methods, including k-means algorithm, on-line
 1852 
     <dd>Convex clustering methods, including k-means algorithm, on-line
1853 
update algorithm (Hard Competitive Learning) and Neural Gas algorithm
 1853 
update algorithm (Hard Competitive Learning) and Neural Gas algorithm
1854 
(Soft Competitive Learning) and calculation of several indexes for
 1854 
(Soft Competitive Learning) and calculation of several indexes for
1855 
finding the number of clusters in a data set.
1855 
finding the number of clusters in a data set.
1856 
<br><dt><strong>cfa</strong>
 1856 
<br><dt><strong>cfa</strong>
1857 
     <dd>Analysis of configuration frequencies.
1857 
     <dd>Analysis of configuration frequencies.
1858 
<br><dt><strong>chron</strong>
 1858 
<br><dt><strong>chron</strong>
1859 
     <dd>A package for working with chronological objects (times and dates).
1859 
     <dd>A package for working with chronological objects (times and dates).
1860 
<br><dt><strong>class</strong>
 1860 
<br><dt><strong>class</strong>
1861 
     <dd>Functions for classification (k-nearest neighbor and LVQ).
1861 
     <dd>Functions for classification (k-nearest neighbor and LVQ).
1862 
Contained in the <code>VR</code> bundle.  <em>Recommended</em>.
1862 
Contained in the <code>VR</code> bundle.  <em>Recommended</em>.
1863 
<br><dt><strong>classPP</strong>
 1863 
<br><dt><strong>classPP</strong>
1864 
     <dd>Projection Pursuit for supervised classification.
1864 
     <dd>Projection Pursuit for supervised classification.
1865 
<br><dt><strong>clim.pact</strong>
 1865 
<br><dt><strong>clim.pact</strong>
1866 
     <dd>Climate analysis and downscaling for monthly and daily data.
1866 
     <dd>Climate analysis and downscaling for monthly and daily data.
1867 
<br><dt><strong>clines</strong>
 1867 
<br><dt><strong>clines</strong>
1868 
     <dd>Calculates Contour Lines.
1868 
     <dd>Calculates Contour Lines.
1869 
<br><dt><strong>cluster</strong>
 1869 
<br><dt><strong>cluster</strong>
1870 
     <dd>Functions for cluster analysis.  <em>Recommended</em>.
1870 
     <dd>Functions for cluster analysis.  <em>Recommended</em>.
1871 
<br><dt><strong>cmprsk</strong>
 1871 
<br><dt><strong>cmprsk</strong>
1872 
     <dd>Estimation, testing and regression modeling of subdistribution functions
 1872 
     <dd>Estimation, testing and regression modeling of subdistribution functions
1873 
in competing risks.
1873 
in competing risks.
1874 
<br><dt><strong>cobs</strong>
 1874 
<br><dt><strong>cobs</strong>
1875 
     <dd>Constrained B-splines: qualitatively constrained (regression) smoothing
 1875 
     <dd>Constrained B-splines: qualitatively constrained (regression) smoothing
1876 
via linear programming.
1876 
via linear programming.
1877 
<br><dt><strong>coda</strong>
 1877 
<br><dt><strong>coda</strong>
1878 
     <dd>Output analysis and diagnostics for Markov Chain Monte Carlo (MCMC)
 1878 
     <dd>Output analysis and diagnostics for Markov Chain Monte Carlo (MCMC)
1879 
simulations.
1879 
simulations.
1880 
<br><dt><strong>combinat</strong>
 1880 
<br><dt><strong>combinat</strong>
1881 
     <dd>Combinatorics utilities.
1881 
     <dd>Combinatorics utilities.
1882 
<br><dt><strong>concord</strong>
 1882 
<br><dt><strong>concord</strong>
1883 
     <dd>Measures of concordance and reliability.
1883 
     <dd>Measures of concordance and reliability.
1884 
<br><dt><strong>conf.design</strong>
 1884 
<br><dt><strong>conf.design</strong>
1885 
     <dd>A series of simple tools for constructing and manipulating confounded
 1885 
     <dd>A series of simple tools for constructing and manipulating confounded
1886 
and fractional factorial designs.
1886 
and fractional factorial designs.
1887 
<br><dt><strong>covRobust</strong>
 1887 
<br><dt><strong>covRobust</strong>
1888 
     <dd>Robust covariance estimation via nearest neighbor cleaning.
1888 
     <dd>Robust covariance estimation via nearest neighbor cleaning.
1889 
<br><dt><strong>cramer</strong>
 1889 
<br><dt><strong>cramer</strong>
1890 
     <dd>Routine for the multivariate nonparametric Cramer test.
1890 
     <dd>Routine for the multivariate nonparametric Cramer test.
1891 
<br><dt><strong>date</strong>
 1891 
<br><dt><strong>date</strong>
1892 
     <dd>Functions for dealing with dates.  The most useful of them accepts a
 1892 
     <dd>Functions for dealing with dates.  The most useful of them accepts a
1893 
vector of input dates in any of the forms <code>8/30/53</code>,
 1893 
vector of input dates in any of the forms <code>8/30/53</code>,
1894 
<code>30Aug53</code>, <code>30 August 1953</code>, <small class="dots">...</small>, <code>August 30 53</code>, or
 1894 
<code>30Aug53</code>, <code>30 August 1953</code>, <small class="dots">...</small>, <code>August 30 53</code>, or
1895 
any mixture of these.
1895 
any mixture of these.
1896 
<br><dt><strong>dblcens</strong>
 1896 
<br><dt><strong>dblcens</strong>
1897 
     <dd>Calculates the NPMLE of the survival distribution for doubly censored
 1897 
     <dd>Calculates the NPMLE of the survival distribution for doubly censored
1898 
data.
1898 
data.
1899 
<br><dt><strong>deal</strong>
 1899 
<br><dt><strong>deal</strong>
1900 
     <dd>Bayesian networks with continuous and/or discrete variables can be
 1900 
     <dd>Bayesian networks with continuous and/or discrete variables can be
1901 
learned and compared from data.
1901 
learned and compared from data.
1902 
<br><dt><strong>debug</strong>
 1902 
<br><dt><strong>debug</strong>
1903 
     <dd>Debugger for R functions, with code display, graceful error recovery,
 1903 
     <dd>Debugger for R functions, with code display, graceful error recovery,
1904 
line-numbered conditional breakpoints, access to exit code, flow
 1904 
line-numbered conditional breakpoints, access to exit code, flow
1905 
control, and full keyboard input.
1905 
control, and full keyboard input.
1906 
<br><dt><strong>deldir</strong>
 1906 
<br><dt><strong>deldir</strong>
1907 
     <dd>Calculates the  Delaunay triangulation and the Dirichlet or Voronoi
 1907 
     <dd>Calculates the  Delaunay triangulation and the Dirichlet or Voronoi
1908 
tesselation (with respect to the entire plane) of a planar point set.
1908 
tesselation (with respect to the entire plane) of a planar point set.
1909 
<br><dt><strong>diamonds</strong>
 1909 
<br><dt><strong>diamonds</strong>
1910 
     <dd>Functions for illustrating aperture-4 diamond partitions in the plane,
 1910 
     <dd>Functions for illustrating aperture-4 diamond partitions in the plane,
1911 
or on the surface of an octahedron or icosahedron, for use as analysis
 1911 
or on the surface of an octahedron or icosahedron, for use as analysis
1912 
or sampling grids.
1912 
or sampling grids.
1913 
<br><dt><strong>dichromat</strong>
 1913 
<br><dt><strong>dichromat</strong>
1914 
     <dd>Color schemes for dichromats: collapse red-green distinctions to
 1914 
     <dd>Color schemes for dichromats: collapse red-green distinctions to
1915 
simulate the effects of colour-blindness.
1915 
simulate the effects of colour-blindness.
1916 
<br><dt><strong>digest</strong>
 1916 
<br><dt><strong>digest</strong>
1917 
     <dd>Two functions for the creation of "hash" digests of arbitrary R
 1917 
     <dd>Two functions for the creation of "hash" digests of arbitrary R
1918 
objects using the md5 and sha-1 algorithms permitting easy comparison of
 1918 
objects using the md5 and sha-1 algorithms permitting easy comparison of
1919 
R language objects.
1919 
R language objects.
1920 
<br><dt><strong>diptest</strong>
 1920 
<br><dt><strong>diptest</strong>
1921 
     <dd>Compute Hartigan's dip test statistic for unimodality.
1921 
     <dd>Compute Hartigan's dip test statistic for unimodality.
1922 
<br><dt><strong>dispmod</strong>
 1922 
<br><dt><strong>dispmod</strong>
1923 
     <dd>Functions for modelling dispersion in GLMs.
1923 
     <dd>Functions for modelling dispersion in GLMs.
1924 
<br><dt><strong>dr</strong>
 1924 
<br><dt><strong>dr</strong>
1925 
     <dd>Functions, methods, and datasets for fitting dimension reduction
 1925 
     <dd>Functions, methods, and datasets for fitting dimension reduction
1926 
regression, including pHd and inverse regression methods SIR and SAVE.
1926 
regression, including pHd and inverse regression methods SIR and SAVE.
1927 
<br><dt><strong>dse</strong>
 1927 
<br><dt><strong>dse</strong>
1928 
     <dd>Dynamic System Estimation, a multivariate time series package.  Contains
 1928 
     <dd>Dynamic System Estimation, a multivariate time series package.  Contains
1929 
<strong>dse1</strong> (the base system, including multivariate ARMA and state
 1929 
<strong>dse1</strong> (the base system, including multivariate ARMA and state
1930 
space models), <strong>dse2</strong> (extensions for evaluating estimation
 1930 
space models), <strong>dse2</strong> (extensions for evaluating estimation
1931 
techniques, forecasting, and for evaluating forecasting model),
 1931 
techniques, forecasting, and for evaluating forecasting model),
1932 
<strong>tframe</strong> (functions for writing code that is independent of the
 1932 
<strong>tframe</strong> (functions for writing code that is independent of the
1933 
representation of time). and <strong>setRNG</strong> (a mechanism for generating
 1933 
representation of time). and <strong>setRNG</strong> (a mechanism for generating
1934 
the same random numbers in S and R).
1934 
the same random numbers in S and R).
1935 
<br><dt><strong>dynamicGraph</strong>
 1935 
<br><dt><strong>dynamicGraph</strong>
1936 
     <dd>Interactive graphical tool for manipulating graphs.
1936 
     <dd>Interactive graphical tool for manipulating graphs.
1937 
<br><dt><strong>e1071</strong>
 1937 
<br><dt><strong>e1071</strong>
1938 
     <dd>Miscellaneous functions used at the Department of Statistics at TU Wien
 1938 
     <dd>Miscellaneous functions used at the Department of Statistics at TU Wien
1939 
(E1071), including moments, short-time Fourier transforms, Independent
 1939 
(E1071), including moments, short-time Fourier transforms, Independent
1940 
Component Analysis, Latent Class Analysis, support vector machines, and
 1940 
Component Analysis, Latent Class Analysis, support vector machines, and
1941 
fuzzy clustering, shortest path computation, bagged clustering, and some
 1941 
fuzzy clustering, shortest path computation, bagged clustering, and some
1942 
more.
1942 
more.
1943 
<br><dt><strong>effects</strong>
 1943 
<br><dt><strong>effects</strong>
1944 
     <dd>Graphical and tabular effect displays, e.g., of interactions, for linear
 1944 
     <dd>Graphical and tabular effect displays, e.g., of interactions, for linear
1945 
and generalised linear models.
1945 
and generalised linear models.
1946 
<br><dt><strong>eha</strong>
 1946 
<br><dt><strong>eha</strong>
1947 
     <dd>A package for survival and event history analysis.
1947 
     <dd>A package for survival and event history analysis.
1948 
<br><dt><strong>ellipse</strong>
 1948 
<br><dt><strong>ellipse</strong>
1949 
     <dd>Package for drawing ellipses and ellipse-like confidence regions.
1949 
     <dd>Package for drawing ellipses and ellipse-like confidence regions.
1950 
<br><dt><strong>emme2</strong>
 1950 
<br><dt><strong>emme2</strong>
1951 
     <dd>Functions to read from and write to an EMME/2 databank.
1951 
     <dd>Functions to read from and write to an EMME/2 databank.
1952 
<br><dt><strong>emplik</strong>
 1952 
<br><dt><strong>emplik</strong>
1953 
     <dd>Empirical likelihood ratio for means/quantiles/hazards from possibly
 1953 
     <dd>Empirical likelihood ratio for means/quantiles/hazards from possibly
1954 
right censored data.
1954 
right censored data.
1955 
<br><dt><strong>energy</strong>
 1955 
<br><dt><strong>energy</strong>
1956 
     <dd>E-statistics (energy) tests for comparing distributions: multivariate
 1956 
     <dd>E-statistics (energy) tests for comparing distributions: multivariate
1957 
normality, Poisson test, multivariate k-sample test for equal
 1957 
normality, Poisson test, multivariate k-sample test for equal
1958 
distributions, hierarchical clustering by e-distances.
1958 
distributions, hierarchical clustering by e-distances.
1959 
<br><dt><strong>evd</strong>
 1959 
<br><dt><strong>evd</strong>
1960 
     <dd>Functions for extreme value distributions.  Extends simulation,
 1960 
     <dd>Functions for extreme value distributions.  Extends simulation,
1961 
distribution, quantile and density functions to univariate, bivariate
 1961 
distribution, quantile and density functions to univariate, bivariate
1962 
and (for simulation) multivariate parametric extreme value
 1962 
and (for simulation) multivariate parametric extreme value
1963 
distributions, and provides fitting functions which calculate maximum
 1963 
distributions, and provides fitting functions which calculate maximum
1964 
likelihood estimates for univariate and bivariate models.
1964 
likelihood estimates for univariate and bivariate models.
1965 
<br><dt><strong>exactLoglinTest</strong>
 1965 
<br><dt><strong>exactLoglinTest</strong>
1966 
     <dd>Monte Carlo exact tests for log-linear models.
1966 
     <dd>Monte Carlo exact tests for log-linear models.
1967 
<br><dt><strong>exactRankTests</strong>
 1967 
<br><dt><strong>exactRankTests</strong>
1968 
     <dd>Computes exact p-values and quantiles using an implementation of
 1968 
     <dd>Computes exact p-values and quantiles using an implementation of
1969 
the Streitberg/Roehmel shift algorithm.
1969 
the Streitberg/Roehmel shift algorithm.
1970 
<br><dt><strong>fastICA</strong>
 1970 
<br><dt><strong>fastICA</strong>
1971 
     <dd>Implementation of FastICA algorithm to perform Independent Component
 1971 
     <dd>Implementation of FastICA algorithm to perform Independent Component
1972 
Analysis (ICA) and Projection Pursuit.
1972 
Analysis (ICA) and Projection Pursuit.
1973 
<br><dt><strong>fda</strong>
 1973 
<br><dt><strong>fda</strong>
1974 
     <dd>Functional Data Analysis: analysis of data where the basic observation
 1974 
     <dd>Functional Data Analysis: analysis of data where the basic observation
1975 
is a function of some sort.
1975 
is a function of some sort.
1976 
<br><dt><strong>fdim</strong>
 1976 
<br><dt><strong>fdim</strong>
1977 
     <dd>Functions for calculating fractal dimension.
1977 
     <dd>Functions for calculating fractal dimension.
1978 
<br><dt><strong>fields</strong>
 1978 
<br><dt><strong>fields</strong>
1979 
     <dd>A collection of programs for curve and function fitting with an emphasis
 1979 
     <dd>A collection of programs for curve and function fitting with an emphasis
1980 
on spatial data.  The major methods implemented include cubic and thin
 1980 
on spatial data.  The major methods implemented include cubic and thin
1981 
plate splines, universal Kriging and Kriging for large data sets.  The
 1981 
plate splines, universal Kriging and Kriging for large data sets.  The
1982 
main feature is that any covariance function implemented in R can be
 1982 
main feature is that any covariance function implemented in R can be
1983 
used for spatial prediction.
1983 
used for spatial prediction.
1984 
<br><dt><strong>flexmix</strong>
 1984 
<br><dt><strong>flexmix</strong>
1985 
     <dd>Flexible Mixture Modeling: a general framework for finite mixtures of
 1985 
     <dd>Flexible Mixture Modeling: a general framework for finite mixtures of
1986 
regression models using the EM algorithm.
1986 
regression models using the EM algorithm.
1987 
<br><dt><strong>foreign</strong>
 1987 
<br><dt><strong>foreign</strong>
1988 
     <dd>Functions for reading and writing data stored by statistical software
 1988 
     <dd>Functions for reading and writing data stored by statistical software
1989 
like Minitab, SAS, SPSS, Stata, etc.  <em>Recommended</em>.
1989 
like Minitab, SAS, SPSS, Stata, etc.  <em>Recommended</em>.
1990 
<br><dt><strong>fork</strong>
 1990 
<br><dt><strong>fork</strong>
1991 
     <dd>Functions for handling multiple processes: simple wrappers around the
 1991 
     <dd>Functions for handling multiple processes: simple wrappers around the
1992 
Unix process management API calls.
1992 
Unix process management API calls.
1993 
<br><dt><strong>forward</strong>
 1993 
<br><dt><strong>forward</strong>
1994 
     <dd>Forward search approach to robust analysis in linear and generalized
 1994 
     <dd>Forward search approach to robust analysis in linear and generalized
1995 
linear regression models.
1995 
linear regression models.
1996 
<br><dt><strong>fpc</strong>
 1996 
<br><dt><strong>fpc</strong>
1997 
     <dd>Fixed point clusters, clusterwise regression and discriminant plots.
1997 
     <dd>Fixed point clusters, clusterwise regression and discriminant plots.
1998 
<br><dt><strong>fracdiff</strong>
 1998 
<br><dt><strong>fracdiff</strong>
1999 
     <dd>Maximum likelihood estimation of the parameters of a fractionally
 1999 
     <dd>Maximum likelihood estimation of the parameters of a fractionally
2000 
differenced ARIMA(p,d,q) model (Haslett and Raftery, Applied
 2000 
differenced ARIMA(p,d,q) model (Haslett and Raftery, Applied
2001 
Statistics, 1989).
2001 
Statistics, 1989).
2002 
<br><dt><strong>ftnonpar</strong>
 2002 
<br><dt><strong>ftnonpar</strong>
2003 
     <dd>Features and strings for nonparametric regression.
2003 
     <dd>Features and strings for nonparametric regression.
2004 
<br><dt><strong>g.data</strong>
 2004 
<br><dt><strong>g.data</strong>
2005 
     <dd>Create and maintain delayed-data packages (DDP's).
2005 
     <dd>Create and maintain delayed-data packages (DDP's).
2006 
<br><dt><strong>gafit</strong>
 2006 
<br><dt><strong>gafit</strong>
2007 
     <dd>Genetic algorithm for curve fitting.
2007 
     <dd>Genetic algorithm for curve fitting.
2008 
<br><dt><strong>gap</strong>
 2008 
<br><dt><strong>gap</strong>
2009 
     <dd>Genetic analysis package for both population and family data.
2009 
     <dd>Genetic analysis package for both population and family data.
2010 
<br><dt><strong>gbm</strong>
 2010 
<br><dt><strong>gbm</strong>
2011 
     <dd>Generalized Boosted Regression Models: implements extensions to Freund
 2011 
     <dd>Generalized Boosted Regression Models: implements extensions to Freund
2012 
and Schapire's AdaBoost algorithm and J. Friedman's gradient boosting
 2012 
and Schapire's AdaBoost algorithm and J. Friedman's gradient boosting
2013 
machine.  Includes regression methods for least squares, absolute loss,
 2013 
machine.  Includes regression methods for least squares, absolute loss,
2014 
logistic, Poisson, Cox proportional hazards partial likelihood, and
 2014 
logistic, Poisson, Cox proportional hazards partial likelihood, and
2015 
AdaBoost exponential loss.
2015 
AdaBoost exponential loss.
2016 
<br><dt><strong>gclus</strong>
 2016 
<br><dt><strong>gclus</strong>
2017 
     <dd>Clustering Graphics.  Orders panels in scatterplot matrices and parallel
 2017 
     <dd>Clustering Graphics.  Orders panels in scatterplot matrices and parallel
2018 
coordinate displays by some merit index.
2018 
coordinate displays by some merit index.
2019 
<br><dt><strong>gee</strong>
 2019 
<br><dt><strong>gee</strong>
2020 
     <dd>An implementation of the Liang/Zeger generalized estimating equation
 2020 
     <dd>An implementation of the Liang/Zeger generalized estimating equation
2021 
approach to GLMs for dependent data.
2021 
approach to GLMs for dependent data.
2022 
<br><dt><strong>geepack</strong>
 2022 
<br><dt><strong>geepack</strong>
2023 
     <dd>Generalized estimating equations solver for parameters in mean, scale,
 2023 
     <dd>Generalized estimating equations solver for parameters in mean, scale,
2024 
and correlation structures, through mean link, scale link, and
 2024 
and correlation structures, through mean link, scale link, and
2025 
correlation link.  Can also handle clustered categorical responses.
2025 
correlation link.  Can also handle clustered categorical responses.
2026 
<br><dt><strong>genetics</strong>
 2026 
<br><dt><strong>genetics</strong>
2027 
     <dd>Classes and methods for handling genetic data.  Includes classes to
 2027 
     <dd>Classes and methods for handling genetic data.  Includes classes to
2028 
represent genotypes and haplotypes at single markers up to multiple
 2028 
represent genotypes and haplotypes at single markers up to multiple
2029 
markers on multiple chromosomes, and functions for allele frequencies,
 2029 
markers on multiple chromosomes, and functions for allele frequencies,
2030 
flagging homo/heterozygotes, flagging carriers of certain alleles,
 2030 
flagging homo/heterozygotes, flagging carriers of certain alleles,
2031 
computing disequlibrium, testing Hardy-Weinberg equilibrium, <small class="dots">...</small>
 2031 
computing disequlibrium, testing Hardy-Weinberg equilibrium, <small class="dots">...</small>
2032 
<br><dt><strong>geoR</strong>
 2032 
<br><dt><strong>geoR</strong>
2033 
     <dd>Functions to perform geostatistical data analysis including model-based
 2033 
     <dd>Functions to perform geostatistical data analysis including model-based
2034 
methods.
2034 
methods.
2035 
<br><dt><strong>geoRglm</strong>
 2035 
<br><dt><strong>geoRglm</strong>
2036 
     <dd>Functions for inference in generalised linear spatial models.
2036 
     <dd>Functions for inference in generalised linear spatial models.
2037 
<br><dt><strong>ggm</strong>
 2037 
<br><dt><strong>ggm</strong>
2038 
     <dd>Functions for defining directed acyclic graphs and undirected graphs,
 2038 
     <dd>Functions for defining directed acyclic graphs and undirected graphs,
2039 
finding induced graphs and fitting Gaussian Markov models.
2039 
finding induced graphs and fitting Gaussian Markov models.
2040 
<br><dt><strong>gld</strong>
 2040 
<br><dt><strong>gld</strong>
2041 
     <dd>Basic functions for the generalised (Tukey) lambda distribution.
2041 
     <dd>Basic functions for the generalised (Tukey) lambda distribution.
2042 
<br><dt><strong>gllm</strong>
 2042 
<br><dt><strong>gllm</strong>
2043 
     <dd>Routines for log-linear models of incomplete contingency tables,
 2043 
     <dd>Routines for log-linear models of incomplete contingency tables,
2044 
including some latent class models via EM and Fisher scoring approaches.
2044 
including some latent class models via EM and Fisher scoring approaches.
2045 
<br><dt><strong>glmmML</strong>
 2045 
<br><dt><strong>glmmML</strong>
2046 
     <dd>A Maximum Likelihood approach to generalized linear models with random
 2046 
     <dd>A Maximum Likelihood approach to generalized linear models with random
2047 
intercept.
2047 
intercept.
2048 
<br><dt><strong>gpclib</strong>
 2048 
<br><dt><strong>gpclib</strong>
2049 
     <dd>General polygon clipping routines for R based on Alan Murta's C
 2049 
     <dd>General polygon clipping routines for R based on Alan Murta's C
2050 
library.
2050 
library.
2051 
<br><dt><strong>grasper</strong>
 2051 
<br><dt><strong>grasper</strong>
2052 
     <dd>Generalized Regression Analysis and Spatial Predictions for R.
2052 
     <dd>Generalized Regression Analysis and Spatial Predictions for R.
2053 
<br><dt><strong>gregmisc</strong>
 2053 
<br><dt><strong>gregmisc</strong>
2054 
     <dd>Miscellaneous functions written/maintained by Gregory R. Warnes.
2054 
     <dd>Miscellaneous functions written/maintained by Gregory R. Warnes.
2055 
<br><dt><strong>gridBase</strong>
 2055 
<br><dt><strong>gridBase</strong>
2056 
     <dd>Integration of base and grid graphics.
2056 
     <dd>Integration of base and grid graphics.
2057 
<br><dt><strong>gss</strong>
 2057 
<br><dt><strong>gss</strong>
2058 
     <dd>A comprehensive package for structural multivariate function estimation
 2058 
     <dd>A comprehensive package for structural multivariate function estimation
2059 
using smoothing splines.
2059 
using smoothing splines.
2060 
<br><dt><strong>gstat</strong>
 2060 
<br><dt><strong>gstat</strong>
2061 
     <dd>multivariable geostatistical modelling, prediction and simulation.
2061 
     <dd>multivariable geostatistical modelling, prediction and simulation.
2062 
Includes code for variogram modelling; simple, ordinary and universal
 2062 
Includes code for variogram modelling; simple, ordinary and universal
2063 
point or block (co)kriging, sequential Gaussian or indicator
 2063 
point or block (co)kriging, sequential Gaussian or indicator
2064 
(co)simulation, and map plotting functions.
2064 
(co)simulation, and map plotting functions.
2065 
<br><dt><strong>gtkDevice</strong>
 2065 
<br><dt><strong>gtkDevice</strong>
2066 
     <dd>GTK graphics device driver that may be used independently of the R-GNOME
 2066 
     <dd>GTK graphics device driver that may be used independently of the R-GNOME
2067 
interface and can be used to create R devices as embedded components in
 2067 
interface and can be used to create R devices as embedded components in
2068 
a GUI using a Gtk drawing area widget, e.g., using RGtk.
2068 
a GUI using a Gtk drawing area widget, e.g., using RGtk.
2069 
<br><dt><strong>hapassoc</strong>
 2069 
<br><dt><strong>hapassoc</strong>
2070 
     <dd>Likelihood inference of trait associations with SNP haplotypes and other
 2070 
     <dd>Likelihood inference of trait associations with SNP haplotypes and other
2071 
attributes using the EM Algorithm.
2071 
attributes using the EM Algorithm.
2072 
<br><dt><strong>haplo.score</strong>
 2072 
<br><dt><strong>haplo.score</strong>
2073 
     <dd>Score tests for association of traits with haplotypes when linkage phase
 2073 
     <dd>Score tests for association of traits with haplotypes when linkage phase
2074 
is ambiguous.
2074 
is ambiguous.
2075 
<br><dt><strong>hdf5</strong>
 2075 
<br><dt><strong>hdf5</strong>
2076 
     <dd>Interface to the <small>NCSA</small> HDF5 library.
2076 
     <dd>Interface to the <small>NCSA</small> HDF5 library.
2077 
<br><dt><strong>hett</strong>
 2077 
<br><dt><strong>hett</strong>
2078 
     <dd>Functions for the fitting and summarizing of heteroscedastic
 2078 
     <dd>Functions for the fitting and summarizing of heteroscedastic
2079 
t-regression.
2079 
t-regression.
2080 
<br><dt><strong>hier.part</strong>
 2080 
<br><dt><strong>hier.part</strong>
2081 
     <dd>Hierarchical Partitioning: variance partition of a multivariate data
 2081 
     <dd>Hierarchical Partitioning: variance partition of a multivariate data
2082 
set.
2082 
set.
2083 
<br><dt><strong>homals</strong>
 2083 
<br><dt><strong>homals</strong>
2084 
     <dd>Homogeneity Analysis (HOMALS) package with optional Tcl/Tk interface.
2084 
     <dd>Homogeneity Analysis (HOMALS) package with optional Tcl/Tk interface.
2085 
<br><dt><strong>hwde</strong>
 2085 
<br><dt><strong>hwde</strong>
2086 
     <dd>Models and tests for departure from Hardy-Weinberg equilibrium and
 2086 
     <dd>Models and tests for departure from Hardy-Weinberg equilibrium and
2087 
independence between loci.
2087 
independence between loci.
2088 
<br><dt><strong>ifs</strong>
 2088 
<br><dt><strong>ifs</strong>
2089 
     <dd>Iterated Function Systems distribution function estimator.
2089 
     <dd>Iterated Function Systems distribution function estimator.
2090 
<br><dt><strong>impute</strong>
 2090 
<br><dt><strong>impute</strong>
2091 
     <dd>Imputation for microarray data (currently KNN only).
2091 
     <dd>Imputation for microarray data (currently KNN only).
2092 
<br><dt><strong>ineq</strong>
 2092 
<br><dt><strong>ineq</strong>
2093 
     <dd>Inequality, concentration and poverty measures, and Lorenz curves
 2093 
     <dd>Inequality, concentration and poverty measures, and Lorenz curves
2094 
(empirical and theoretic).
2094 
(empirical and theoretic).
2095 
<br><dt><strong>ipred</strong>
 2095 
<br><dt><strong>ipred</strong>
2096 
     <dd>Improved predictive models by direct and indirect bootstrap aggregation
 2096 
     <dd>Improved predictive models by direct and indirect bootstrap aggregation
2097 
in classification and regression as well as resampling based estimators
 2097 
in classification and regression as well as resampling based estimators
2098 
of prediction error.
2098 
of prediction error.
2099 
<br><dt><strong>ismev</strong>
 2099 
<br><dt><strong>ismev</strong>
2100 
     <dd>Functions to support the computations carried out in "An Introduction
 2100 
     <dd>Functions to support the computations carried out in "An Introduction
2101 
to Statistical Modeling of Extreme Values;' by S. Coles, 2001, Springer.
2101 
to Statistical Modeling of Extreme Values;' by S. Coles, 2001, Springer.
2102 
The functions may be divided into the following groups; maxima/minima,
 2102 
The functions may be divided into the following groups; maxima/minima,
2103 
order statistics, peaks over thresholds and point processes.
2103 
order statistics, peaks over thresholds and point processes.
2104 
<br><dt><strong>its</strong>
 2104 
<br><dt><strong>its</strong>
2105 
     <dd>An S4 class for handling irregular time series.
2105 
     <dd>An S4 class for handling irregular time series.
2106 
<br><dt><strong>kernlab</strong>
 2106 
<br><dt><strong>kernlab</strong>
2107 
     <dd>Kernel-based machine learning methods including support vector machines.
2107 
     <dd>Kernel-based machine learning methods including support vector machines.
2108 
(Currently in <code>1.9.0/Other</code>.)
2108 
(Currently in <code>1.9.0/Other</code>.)
2109 
<br><dt><strong>knnTree</strong>
 2109 
<br><dt><strong>knnTree</strong>
2110 
     <dd>Construct or predict with k-nearest-neighbor classifiers, using
 2110 
     <dd>Construct or predict with k-nearest-neighbor classifiers, using
2111 
cross-validation to select k, choose variables (by forward or
 2111 
cross-validation to select k, choose variables (by forward or
2112 
backwards selection), and choose scaling (from among no scaling, scaling
 2112 
backwards selection), and choose scaling (from among no scaling, scaling
2113 
each column by its SD, or scaling each column by its MAD).  The finished
 2113 
each column by its SD, or scaling each column by its MAD).  The finished
2114 
classifier will consist of a classification tree with one such
 2114 
classifier will consist of a classification tree with one such
2115 
k-nn classifier in each leaf.
2115 
k-nn classifier in each leaf.
2116 
<br><dt><strong>labstatR</strong>
 2116 
<br><dt><strong>labstatR</strong>
2117 
     <dd>Functions for the book "Laboratorio di statistica con R" by
 2117 
     <dd>Functions for the book "Laboratorio di statistica con R" by
2118 
S. M. Iacus and G. Masarotto, 2002, McGraw-Hill.  Function names and
 2118 
S. M. Iacus and G. Masarotto, 2002, McGraw-Hill.  Function names and
2119 
documentation in Italian.
2119 
documentation in Italian.
2120 
<br><dt><strong>lars</strong>
 2120 
<br><dt><strong>lars</strong>
2121 
     <dd>Least Angle Regression, Lasso and Forward Stagewise: efficient
 2121 
     <dd>Least Angle Regression, Lasso and Forward Stagewise: efficient
2122 
procedures for fitting an entire lasso sequence with the cost of a
 2122 
procedures for fitting an entire lasso sequence with the cost of a
2123 
single least squares fit.
2123 
single least squares fit.
2124 
<br><dt><strong>lasso2</strong>
 2124 
<br><dt><strong>lasso2</strong>
2125 
     <dd>Routines and documentation for solving regression problems while
 2125 
     <dd>Routines and documentation for solving regression problems while
2126 
imposing an L1 constraint on the estimates, based on the algorithm of
 2126 
imposing an L1 constraint on the estimates, based on the algorithm of
2127 
Osborne et al. (1998)
 2127 
Osborne et al. (1998)
2128 
<br><dt><strong>lattice</strong>
 2128 
<br><dt><strong>lattice</strong>
2129 
     <dd>Lattice graphics, an implementation of Trellis Graphics functions.
2129 
     <dd>Lattice graphics, an implementation of Trellis Graphics functions.
2130 
<em>Recommended</em>.
2130 
<em>Recommended</em>.
2131 
<br><dt><strong>lazy</strong>
 2131 
<br><dt><strong>lazy</strong>
2132 
     <dd>Lazy learning for local regression.
2132 
     <dd>Lazy learning for local regression.
2133 
<br><dt><strong>ldDesign</strong>
 2133 
<br><dt><strong>ldDesign</strong>
2134 
     <dd>Design of experiments for detection of linkage disequilibrium,
 2134 
     <dd>Design of experiments for detection of linkage disequilibrium,
2135 
<br><dt><strong>leaps</strong>
 2135 
<br><dt><strong>leaps</strong>
2136 
     <dd>A package which performs an exhaustive search for the best subsets of a
 2136 
     <dd>A package which performs an exhaustive search for the best subsets of a
2137 
given set of potential regressors, using a branch-and-bound algorithm,
 2137 
given set of potential regressors, using a branch-and-bound algorithm,
2138 
and also performs searches using a number of less time-consuming
 2138 
and also performs searches using a number of less time-consuming
2139 
techniques.
2139 
techniques.
2140 
<br><dt><strong>lgtdl</strong>
 2140 
<br><dt><strong>lgtdl</strong>
2141 
     <dd>A set of methods for longitudinal data objects.
2141 
     <dd>A set of methods for longitudinal data objects.
2142 
<br><dt><strong>linprog</strong>
 2142 
<br><dt><strong>linprog</strong>
2143 
     <dd>Solve linear programming/linear optimization problems by using the
 2143 
     <dd>Solve linear programming/linear optimization problems by using the
2144 
simplex algorithm.
2144 
simplex algorithm.
2145 
<br><dt><strong>lme4</strong>
 2145 
<br><dt><strong>lme4</strong>
2146 
     <dd>Fit linear and generalized linear mixed-effects models.
2146 
     <dd>Fit linear and generalized linear mixed-effects models.
2147 
<br><dt><strong>lmeSplines</strong>
 2147 
<br><dt><strong>lmeSplines</strong>
2148 
     <dd>Fit smoothing spline terms in Gaussian linear and nonlinear
 2148 
     <dd>Fit smoothing spline terms in Gaussian linear and nonlinear
2149 
mixed-effects models.
2149 
mixed-effects models.
2150 
<br><dt><strong>lmm</strong>
 2150 
<br><dt><strong>lmm</strong>
2151 
     <dd>Linear mixed models.
2151 
     <dd>Linear mixed models.
2152 
<br><dt><strong>lmtest</strong>
 2152 
<br><dt><strong>lmtest</strong>
2153 
     <dd>A collection of tests on the assumptions of linear regression models
 2153 
     <dd>A collection of tests on the assumptions of linear regression models
2154 
from the book "The linear regression model under test" by W. Kraemer
 2154 
from the book "The linear regression model under test" by W. Kraemer
2155 
and H. Sonnberger, 1986, Physica.
2155 
and H. Sonnberger, 1986, Physica.
2156 
<br><dt><strong>locfit</strong>
 2156 
<br><dt><strong>locfit</strong>
2157 
     <dd>Local Regression, likelihood and density estimation.
2157 
     <dd>Local Regression, likelihood and density estimation.
2158 
<br><dt><strong>logistf</strong>
 2158 
<br><dt><strong>logistf</strong>
2159 
     <dd>Firth's bias reduced logistic regression approach with penalized profile
 2159 
     <dd>Firth's bias reduced logistic regression approach with penalized profile
2160 
likelihood based confidence intervals for parameter estimates.
2160 
likelihood based confidence intervals for parameter estimates.
2161 
<br><dt><strong>logspline</strong>
 2161 
<br><dt><strong>logspline</strong>
2162 
     <dd>Logspline density estimation.
2162 
     <dd>Logspline density estimation.
2163 
<br><dt><strong>lokern</strong>
 2163 
<br><dt><strong>lokern</strong>
2164 
     <dd>Kernel regression smoothing with adaptive local or global plug-in
 2164 
     <dd>Kernel regression smoothing with adaptive local or global plug-in
2165 
bandwidth selection.
2165 
bandwidth selection.
2166 
<br><dt><strong>lpSolve</strong>
 2166 
<br><dt><strong>lpSolve</strong>
2167 
     <dd>Functions that solve general linear/integer problems, assignment
 2167 
     <dd>Functions that solve general linear/integer problems, assignment
2168 
problems, and transportation problems via interfacing Lp_solve.
2168 
problems, and transportation problems via interfacing Lp_solve.
2169 
<br><dt><strong>lpridge</strong>
 2169 
<br><dt><strong>lpridge</strong>
2170 
     <dd>Local polynomial (ridge) regression.
2170 
     <dd>Local polynomial (ridge) regression.
2171 
<br><dt><strong>magic</strong>
 2171 
<br><dt><strong>magic</strong>
2172 
     <dd>A variety of methods for creating magic squares of any order greater
 2172 
     <dd>A variety of methods for creating magic squares of any order greater
2173 
than 2, and various magic hypercubes.
2173 
than 2, and various magic hypercubes.
2174 
<br><dt><strong>mapdata</strong>
 2174 
<br><dt><strong>mapdata</strong>
2175 
     <dd>Supplement to package <strong>maps</strong>, providing the larger and/or
 2175 
     <dd>Supplement to package <strong>maps</strong>, providing the larger and/or
2176 
higher-resolution databases.
2176 
higher-resolution databases.
2177 
<br><dt><strong>mapproj</strong>
 2177 
<br><dt><strong>mapproj</strong>
2178 
     <dd>Map Projections: converts latitude/longitude into projected coordinates.
2178 
     <dd>Map Projections: converts latitude/longitude into projected coordinates.
2179 
<br><dt><strong>maps</strong>
 2179 
<br><dt><strong>maps</strong>
2180 
     <dd>Draw geographical maps.  Projection code and larger maps are in separate
 2180 
     <dd>Draw geographical maps.  Projection code and larger maps are in separate
2181 
packages.
2181 
packages.
2182 
<br><dt><strong>maptools</strong>
 2182 
<br><dt><strong>maptools</strong>
2183 
     <dd>Set of tools for manipulating and reading geographic data, in particular
 2183 
     <dd>Set of tools for manipulating and reading geographic data, in particular
2184 
ESRI shapefiles.
2184 
ESRI shapefiles.
2185 
<br><dt><strong>maptree</strong>
 2185 
<br><dt><strong>maptree</strong>
2186 
     <dd>Functions with example data for graphing and mapping models from
 2186 
     <dd>Functions with example data for graphing and mapping models from
2187 
hierarchical clustering and classification and regression trees.
2187 
hierarchical clustering and classification and regression trees.
2188 
<br><dt><strong>maxstat</strong>
 2188 
<br><dt><strong>maxstat</strong>
2189 
     <dd>Maximally selected rank and Gauss statistics with several p-value
 2189 
     <dd>Maximally selected rank and Gauss statistics with several p-value
2190 
approximations.
2190 
approximations.
2191 
<br><dt><strong>mclust</strong>
 2191 
<br><dt><strong>mclust</strong>
2192 
     <dd>Model-based cluster analysis: the 2002 version of MCLUST.
2192 
     <dd>Model-based cluster analysis: the 2002 version of MCLUST.
2193 
<br><dt><strong>mclust1998</strong>
 - 
 
2194 
     <dd>Model-based cluster analysis: the 1998 version of MCLUST.
- 
 
2195 
<br><dt><strong>mda</strong>
 2193 
<br><dt><strong>mda</strong>
2196 
     <dd>Code for mixture discriminant analysis (MDA), flexible discriminant
 2194 
     <dd>Code for mixture discriminant analysis (MDA), flexible discriminant
2197 
analysis (FDA), penalized discriminant analysis (PDA), multivariate
 2195 
analysis (FDA), penalized discriminant analysis (PDA), multivariate
2198 
additive regression splines (MARS), adaptive back-fitting splines
 2196 
additive regression splines (MARS), adaptive back-fitting splines
2199 
(BRUTO), and penalized regression.
2197 
(BRUTO), and penalized regression.
2200 
<br><dt><strong>meanscore</strong>
 2198 
<br><dt><strong>meanscore</strong>
2201 
     <dd>Mean Score method for missing covariate data in logistic regression
 2199 
     <dd>Mean Score method for missing covariate data in logistic regression
2202 
models.
2200 
models.
2203 
<br><dt><strong>merror</strong>
 2201 
<br><dt><strong>merror</strong>
2204 
     <dd>Accuracy and precision of measurements.
2202 
     <dd>Accuracy and precision of measurements.
2205 
<br><dt><strong>mgcv</strong>
 2203 
<br><dt><strong>mgcv</strong>
2206 
     <dd>Routines for GAMs and other genralized ridge regression problems with
 2204 
     <dd>Routines for GAMs and other genralized ridge regression problems with
2207 
multiple smoothing parameter selection by GCV or UBRE.
2205 
multiple smoothing parameter selection by GCV or UBRE.
2208 
<em>Recommended</em>.
2206 
<em>Recommended</em>.
2209 
<br><dt><strong>mimR</strong>
 2207 
<br><dt><strong>mimR</strong>
2210 
     <dd>An R interface to MIM for graphical modeling in R.
2208 
     <dd>An R interface to MIM for graphical modeling in R.
2211 
<br><dt><strong>mix</strong>
 2209 
<br><dt><strong>mix</strong>
2212 
     <dd>Estimation/multiple imputation programs for mixed categorical and
 2210 
     <dd>Estimation/multiple imputation programs for mixed categorical and
2213 
continuous data.
2211 
continuous data.
2214 
<br><dt><strong>mlbench</strong>
 2212 
<br><dt><strong>mlbench</strong>
2215 
     <dd>A collection of artificial and real-world machine learning benchmark
 2213 
     <dd>A collection of artificial and real-world machine learning benchmark
2216 
problems, including the Boston housing data.
2214 
problems, including the Boston housing data.
2217 
<br><dt><strong>mmlcr</strong>
 2215 
<br><dt><strong>mmlcr</strong>
2218 
     <dd>Mixed-mode latent class regression (also known as mixed-mode mixture
 2216 
     <dd>Mixed-mode latent class regression (also known as mixed-mode mixture
2219 
model regression or mixed-mode mixture regression models) which can
 2217 
model regression or mixed-mode mixture regression models) which can
2220 
handle both longitudinal and one-time responses.
2218 
handle both longitudinal and one-time responses.
2221 
<br><dt><strong>moc</strong>
 2219 
<br><dt><strong>moc</strong>
2222 
     <dd>Fits a variety of mixtures models for multivariate observations with
 2220 
     <dd>Fits a variety of mixtures models for multivariate observations with
2223 
user-difined distributions and curves.
2221 
user-difined distributions and curves.
2224 
<br><dt><strong>mscalib</strong>
 2222 
<br><dt><strong>mscalib</strong>
2225 
     <dd>Calibration and filtering of MALDI-TOF Peptide Mass Fingerprint data.
2223 
     <dd>Calibration and filtering of MALDI-TOF Peptide Mass Fingerprint data.
2226 
<br><dt><strong>msm</strong>
 2224 
<br><dt><strong>msm</strong>
2227 
     <dd>Functions for fitting continuous-time Markov multi-state models to
 2225 
     <dd>Functions for fitting continuous-time Markov multi-state models to
2228 
categorical processes observed at arbitrary times, optionally with
 2226 
categorical processes observed at arbitrary times, optionally with
2229 
misclassified responses, and covariates on transition or
 2227 
misclassified responses, and covariates on transition or
2230 
misclassification rates.
2228 
misclassification rates.
2231 
<br><dt><strong>muhaz</strong>
 2229 
<br><dt><strong>muhaz</strong>
2232 
     <dd>Hazard function estimation in survival analysis.
2230 
     <dd>Hazard function estimation in survival analysis.
2233 
<br><dt><strong>multcomp</strong>
 2231 
<br><dt><strong>multcomp</strong>
2234 
     <dd>Multiple comparison procedures for the one-way layout.
2232 
     <dd>Multiple comparison procedures for the one-way layout.
2235 
<br><dt><strong>multidim</strong>
 2233 
<br><dt><strong>multidim</strong>
2236 
     <dd>Multidimensional descriptive statistics: factorial methods and
 2234 
     <dd>Multidimensional descriptive statistics: factorial methods and
2237 
classification.
2235 
classification.
- 
 
 2236 
<br><dt><strong>multinomRob</strong>
- 
 
 2237 
     <dd>Overdispersed multinomial regression using robust (LQD and tanh)
- 
 
 2238 
estimation.
2238 
<br><dt><strong>multiv</strong>
 2239 
<br><dt><strong>multiv</strong>
2239 
     <dd>Functions for hierarchical clustering, partitioning, bond energy
 2240 
     <dd>Functions for hierarchical clustering, partitioning, bond energy
2240 
algorithm, Sammon mapping, PCA and correspondence analysis.
2241 
algorithm, Sammon mapping, PCA and correspondence analysis.
2241 
<br><dt><strong>mvbutils</strong>
 2242 
<br><dt><strong>mvbutils</strong>
2242 
     <dd>Utilities by Mark V. Bravington for project organization, editing and
 2243 
     <dd>Utilities by Mark V. Bravington for project organization, editing and
2243 
backup, sourcing, documentation (formal and informal), package
 2244 
backup, sourcing, documentation (formal and informal), package
2244 
preparation, macro functions, and more.
2245 
preparation, macro functions, and more.
2245 
<br><dt><strong>mvnmle</strong>
 2246 
<br><dt><strong>mvnmle</strong>
2246 
     <dd>ML estimation for multivariate normal data with missing values.
2247 
     <dd>ML estimation for multivariate normal data with missing values.
2247 
<br><dt><strong>mvnormtest</strong>
 2248 
<br><dt><strong>mvnormtest</strong>
2248 
     <dd>Generalization of the Shapiro-Wilk test for multivariate variables.
2249 
     <dd>Generalization of the Shapiro-Wilk test for multivariate variables.
2249 
<br><dt><strong>mvpart</strong>
 2250 
<br><dt><strong>mvpart</strong>
2250 
     <dd>Multivariate partitioning.
2251 
     <dd>Multivariate partitioning.
2251 
<br><dt><strong>mvtnorm</strong>
 2252 
<br><dt><strong>mvtnorm</strong>
2252 
     <dd>Multivariate normal and t distributions.
2253 
     <dd>Multivariate normal and t distributions.
2253 
<br><dt><strong>ncdf</strong>
 2254 
<br><dt><strong>ncdf</strong>
2254 
     <dd>Interface to Unidata netCDF data files.
2255 
     <dd>Interface to Unidata netCDF data files.
2255 
<br><dt><strong>ncomplete</strong>
 2256 
<br><dt><strong>ncomplete</strong>
2256 
     <dd>Functions to perform the regression depth method (RDM) to binary
 2257 
     <dd>Functions to perform the regression depth method (RDM) to binary
2257 
regression to approximate the minimum number of observations that can be
 2258 
regression to approximate the minimum number of observations that can be
2258 
removed such that the reduced data set has complete separation.
2259 
removed such that the reduced data set has complete separation.
2259 
<br><dt><strong>negenes</strong>
 2260 
<br><dt><strong>negenes</strong>
2260 
     <dd>Estimating the number of essential genes in a genome on the basis of
 2261 
     <dd>Estimating the number of essential genes in a genome on the basis of
2261 
data from a random transposon mutagenesis experiment, through the use of
 2262 
data from a random transposon mutagenesis experiment, through the use of
2262 
a Gibbs sampler.
2263 
a Gibbs sampler.
2263 
<br><dt><strong>netCDF</strong>
 2264 
<br><dt><strong>netCDF</strong>
2264 
     <dd>Read data from netCDF files.
2265 
     <dd>Read data from netCDF files.
2265 
<br><dt><strong>nlme</strong>
 2266 
<br><dt><strong>nlme</strong>
2266 
     <dd>Fit and compare Gaussian linear and nonlinear mixed-effects models.
2267 
     <dd>Fit and compare Gaussian linear and nonlinear mixed-effects models.
2267 
<em>Recommended</em>.
2268 
<em>Recommended</em>.
2268 
<br><dt><strong>nlmeODE</strong>
 2269 
<br><dt><strong>nlmeODE</strong>
2269 
     <dd>Combine the <strong>nlme</strong> and <strong>odesolve</strong> packages for
 2270 
     <dd>Combine the <strong>nlme</strong> and <strong>odesolve</strong> packages for
2270 
mixed-effects modelling using differential equations.
2271 
mixed-effects modelling using differential equations.
2271 
<br><dt><strong>nlrq</strong>
 2272 
<br><dt><strong>nlrq</strong>
2272 
     <dd>Nonlinear quantile regression.
2273 
     <dd>Nonlinear quantile regression.
2273 
<br><dt><strong>nnet</strong>
 2274 
<br><dt><strong>nnet</strong>
2274 
     <dd>Software for single hidden layer perceptrons ("feed-forward neural
 2275 
     <dd>Software for single hidden layer perceptrons ("feed-forward neural
2275 
networks"), and for multinomial log-linear models.  Contained in the
 2276 
networks"), and for multinomial log-linear models.  Contained in the
2276 
<code>VR</code> bundle.  <em>Recommended</em>.
2277 
<code>VR</code> bundle.  <em>Recommended</em>.
2277 
<br><dt><strong>nor1mix</strong>
 2278 
<br><dt><strong>nor1mix</strong>
2278 
     <dd>One-dimensional normal mixture models classes, for, e.g., density
 2279 
     <dd>One-dimensional normal mixture models classes, for, e.g., density
2279 
estimation or clustering algorithms research and teaching; providing the
 2280 
estimation or clustering algorithms research and teaching; providing the
2280 
widely used Marron-Wand densities.
2281 
widely used Marron-Wand densities.
2281 
<br><dt><strong>norm</strong>
 2282 
<br><dt><strong>norm</strong>
2282 
     <dd>Analysis of multivariate normal datasets with missing values.
2283 
     <dd>Analysis of multivariate normal datasets with missing values.
2283 
<br><dt><strong>normalp</strong>
 2284 
<br><dt><strong>normalp</strong>
2284 
     <dd>A collection of utilities for normal of order p distributions
 2285 
     <dd>A collection of utilities for normal of order p distributions
2285 
(General Error Distributions).
2286 
(General Error Distributions).
2286 
<br><dt><strong>nortest</strong>
 2287 
<br><dt><strong>nortest</strong>
2287 
     <dd>Five omnibus tests for the composite hypothesis of normality.
2288 
     <dd>Five omnibus tests for the composite hypothesis of normality.
2288 
<br><dt><strong>noverlap</strong>
 2289 
<br><dt><strong>noverlap</strong>
2289 
     <dd>Functions to perform the regression depth method (RDM) to binary
 2290 
     <dd>Functions to perform the regression depth method (RDM) to binary
2290 
regression to approximate the amount of overlap, i.e., the minimal
 2291 
regression to approximate the amount of overlap, i.e., the minimal
2291 
number of observations that need to be removed such that the reduced
 2292 
number of observations that need to be removed such that the reduced
2292 
data set has no longer overlap.
2293 
data set has no longer overlap.
2293 
<br><dt><strong>npmc</strong>
 2294 
<br><dt><strong>npmc</strong>
2294 
     <dd>Nonparametric Multiple Comparisons:  provides simultaneous rank test
 2295 
     <dd>Nonparametric Multiple Comparisons:  provides simultaneous rank test
2295 
procedures for the one-way layout without presuming a certain
 2296 
procedures for the one-way layout without presuming a certain
2296 
distribution.
2297 
distribution.
2297 
<br><dt><strong>nprq</strong>
 2298 
<br><dt><strong>nprq</strong>
2298 
     <dd>Nonparametric and sparse quantile regression methods.
2299 
     <dd>Nonparametric and sparse quantile regression methods.
2299 
<br><dt><strong>odesolve</strong>
 2300 
<br><dt><strong>odesolve</strong>
2300 
     <dd>An interface for the Ordinary Differential Equation (ODE) solver lsoda.
2301 
     <dd>An interface for the Ordinary Differential Equation (ODE) solver lsoda.
2301 
ODEs are expressed as R functions.
2302 
ODEs are expressed as R functions.
2302 
<br><dt><strong>orientlib</strong>
 2303 
<br><dt><strong>orientlib</strong>
2303 
     <dd>Representations, conversions and display of orientation SO(3) data.
2304 
     <dd>Representations, conversions and display of orientation SO(3) data.
2304 
<br><dt><strong>oz</strong>
 2305 
<br><dt><strong>oz</strong>
2305 
     <dd>Functions for plotting Australia's coastline and state boundaries.
2306 
     <dd>Functions for plotting Australia's coastline and state boundaries.
2306 
<br><dt><strong>pamr</strong>
 2307 
<br><dt><strong>pamr</strong>
2307 
     <dd>Pam: Prediction Analysis for Microarrays.
2308 
     <dd>Pam: Prediction Analysis for Microarrays.
2308 
<br><dt><strong>pan</strong>
 2309 
<br><dt><strong>pan</strong>
2309 
     <dd>Multiple imputation for multivariate panel or clustered data.
2310 
     <dd>Multiple imputation for multivariate panel or clustered data.
2310 
<br><dt><strong>panel</strong>
 2311 
<br><dt><strong>panel</strong>
2311 
     <dd>Functions and datasets for fitting models to Panel data.
2312 
     <dd>Functions and datasets for fitting models to Panel data.
2312 
<br><dt><strong>pastecs</strong>
 2313 
<br><dt><strong>pastecs</strong>
2313 
     <dd>Package for Analysis of Space-Time Ecological Series.
2314 
     <dd>Package for Analysis of Space-Time Ecological Series.
2314 
<br><dt><strong>pcurve</strong>
 2315 
<br><dt><strong>pcurve</strong>
2315 
     <dd>Fits a principal curve to a numeric multivariate dataset in arbitrary
 2316 
     <dd>Fits a principal curve to a numeric multivariate dataset in arbitrary
2316 
dimensions.  Produces diagnostic plots.  Also calculates Bray-Curtis and
 2317 
dimensions.  Produces diagnostic plots.  Also calculates Bray-Curtis and
2317 
other distance matrices and performs multi-dimensional scaling and
 2318 
other distance matrices and performs multi-dimensional scaling and
2318 
principal component analyses.
2319 
principal component analyses.
2319 
<br><dt><strong>pear</strong>
 2320 
<br><dt><strong>pear</strong>
2320 
     <dd>Periodic Autoregression Analysis.
2321 
     <dd>Periodic Autoregression Analysis.
2321 
<br><dt><strong>permax</strong>
 2322 
<br><dt><strong>permax</strong>
2322 
     <dd>Functions intended to facilitate certain basic analyses of DNA array
 2323 
     <dd>Functions intended to facilitate certain basic analyses of DNA array
2323 
data, especially with regard to comparing expression levels between two
 2324 
data, especially with regard to comparing expression levels between two
2324 
types of tissue.
2325 
types of tissue.
2325 
<br><dt><strong>pheno</strong>
 2326 
<br><dt><strong>pheno</strong>
2326 
     <dd>Some easy-to-use functions for time series analyses of (plant-)
 2327 
     <dd>Some easy-to-use functions for time series analyses of (plant-)
2327 
phenological data sets.
2328 
phenological data sets.
2328 
<br><dt><strong>phyloarray</strong>
 2329 
<br><dt><strong>phyloarray</strong>
2329 
     <dd>Software to process data from phylogenetic or identification
 2330 
     <dd>Software to process data from phylogenetic or identification
2330 
microarrays.
2331 
microarrays.
2331 
<br><dt><strong>pinktoe</strong>
 2332 
<br><dt><strong>pinktoe</strong>
2332 
     <dd>Converts S trees to <small>HTML</small>/Perl files for interactive tree traversal.
2333 
     <dd>Converts S trees to <small>HTML</small>/Perl files for interactive tree traversal.
2333 
<br><dt><strong>pixmap</strong>
 2334 
<br><dt><strong>pixmap</strong>
2334 
     <dd>Functions for import, export, plotting and other manipulations of
 2335 
     <dd>Functions for import, export, plotting and other manipulations of
2335 
bitmapped images.
2336 
bitmapped images.
2336 
<br><dt><strong>pls.pcr</strong>
 2337 
<br><dt><strong>pls.pcr</strong>
2337 
     <dd>Multivariate regression by PLS and PCR.
2338 
     <dd>Multivariate regression by PLS and PCR.
2338 
<br><dt><strong>polspline</strong>
 2339 
<br><dt><strong>polspline</strong>
2339 
     <dd>Routines for the polynomial spline fitting routines hazard regression,
 2340 
     <dd>Routines for the polynomial spline fitting routines hazard regression,
2340 
hazard estimation with flexible tails, logspline, lspec, polyclass, and
 2341 
hazard estimation with flexible tails, logspline, lspec, polyclass, and
2341 
polymars, by C. Kooperberg and co-authors.
2342 
polymars, by C. Kooperberg and co-authors.
2342 
<br><dt><strong>polynom</strong>
 2343 
<br><dt><strong>polynom</strong>
2343 
     <dd>A collection of functions to implement a class for univariate polynomial
 2344 
     <dd>A collection of functions to implement a class for univariate polynomial
2344 
manipulations.
2345 
manipulations.
2345 
<br><dt><strong>pps</strong>
 2346 
<br><dt><strong>pps</strong>
2346 
     <dd>Functions to select samples using PPS (probability proportional to size)
 2347 
     <dd>Functions to select samples using PPS (probability proportional to size)
2347 
sampling, for stratified simple random sampling, and to compute joint
 2348 
sampling, for stratified simple random sampling, and to compute joint
2348 
inclusion probabilities for Sampford's method of PPS sampling.
2349 
inclusion probabilities for Sampford's method of PPS sampling.
2349 
<br><dt><strong>prabclus</strong>
 2350 
<br><dt><strong>prabclus</strong>
2350 
     <dd>Distance based parametric bootstrap tests for clustering, mainly thought
 2351 
     <dd>Distance based parametric bootstrap tests for clustering, mainly thought
2351 
for presence-absence data (clustering of species distribution maps).
2352 
for presence-absence data (clustering of species distribution maps).
2352 
Jaccard and Kulczynski distance measures, clustering of MDS scores, and
 2353 
Jaccard and Kulczynski distance measures, clustering of MDS scores, and
2353 
nearest neighbor based noise detection.
2354 
nearest neighbor based noise detection.
2354 
<br><dt><strong>princurve</strong>
 2355 
<br><dt><strong>princurve</strong>
2355 
     <dd>Fits a principal curve to a matrix of points in arbitrary dimension.
2356 
     <dd>Fits a principal curve to a matrix of points in arbitrary dimension.
2356 
<br><dt><strong>pspline</strong>
 2357 
<br><dt><strong>pspline</strong>
2357 
     <dd>Smoothing splines with penalties on order m derivatives.
2358 
     <dd>Smoothing splines with penalties on order m derivatives.
2358 
<br><dt><strong>psy</strong>
 2359 
<br><dt><strong>psy</strong>
2359 
     <dd>Various procedures used in psychometry: Kappa, ICC, Cronbach alpha,
 2360 
     <dd>Various procedures used in psychometry: Kappa, ICC, Cronbach alpha,
2360 
screeplot, PCA and related methods.
2361 
screeplot, PCA and related methods.
2361 
<br><dt><strong>qtl</strong>
 2362 
<br><dt><strong>qtl</strong>
2362 
     <dd>Analysis of experimental crosses to identify QTLs.
2363 
     <dd>Analysis of experimental crosses to identify QTLs.
2363 
<br><dt><strong>quadprog</strong>
 2364 
<br><dt><strong>quadprog</strong>
2364 
     <dd>For solving quadratic programming problems.
2365 
     <dd>For solving quadratic programming problems.
2365 
<br><dt><strong>quantreg</strong>
 2366 
<br><dt><strong>quantreg</strong>
2366 
     <dd>Quantile regression and related methods.
2367 
     <dd>Quantile regression and related methods.
2367 
<br><dt><strong>qvcalc</strong>
 2368 
<br><dt><strong>qvcalc</strong>
2368 
     <dd>Functions to compute quasi-variances and associated measures of
 2369 
     <dd>Functions to compute quasi-variances and associated measures of
2369 
approximation error.
2370 
approximation error.
2370 
<br><dt><strong>randomForest</strong>
 2371 
<br><dt><strong>randomForest</strong>
2371 
     <dd>Breiman's random forest classifier.
2372 
     <dd>Breiman's random forest classifier.
- 
 
 2373 
<br><dt><strong>ref</strong>
- 
 
 2374 
     <dd>Functions for creating references, reading from and writing ro
- 
 
 2375 
references and a memory efficient refdata type that transparently
- 
 
 2376 
encapsulates matrices and data frames.
2372 
<br><dt><strong>relimp</strong>
 2377 
<br><dt><strong>relimp</strong>
2373 
     <dd>Functions to facilitate inference on the relative importance of
 2378 
     <dd>Functions to facilitate inference on the relative importance of
2374 
predictors in a linear or generalized linear model.
2379 
predictors in a linear or generalized linear model.
2375 
<br><dt><strong>rgdal</strong>
 2380 
<br><dt><strong>rgdal</strong>
2376 
     <dd>Provides bindings to Frank Warmerdam's Geospatial Data Abstraction
 2381 
     <dd>Provides bindings to Frank Warmerdam's Geospatial Data Abstraction
2377 
Library (GDAL).
2382 
Library (GDAL).
2378 
<br><dt><strong>rgenoud</strong>
 2383 
<br><dt><strong>rgenoud</strong>
2379 
     <dd>R version of GENetic Optimization Using Derivatives.
2384 
     <dd>R version of GENetic Optimization Using Derivatives.
2380 
<br><dt><strong>rimage</strong>
 2385 
<br><dt><strong>rimage</strong>
2381 
     <dd>Functions for image processing, including Sobel filter, rank filters,
 2386 
     <dd>Functions for image processing, including Sobel filter, rank filters,
2382 
fft, histogram equalization, and reading <small>JPEG</small> files.
2387 
fft, histogram equalization, and reading <small>JPEG</small> files.
2383 
<br><dt><strong>rmeta</strong>
 2388 
<br><dt><strong>rmeta</strong>
2384 
     <dd>Functions for simple fixed and random effects meta-analysis for
 2389 
     <dd>Functions for simple fixed and random effects meta-analysis for
2385 
two-sample comparison of binary outcomes.
2390 
two-sample comparison of binary outcomes.
2386 
<br><dt><strong>rpart</strong>
 2391 
<br><dt><strong>rpart</strong>
2387 
     <dd>Recursive PARTitioning and regression trees.  <em>Recommended</em>.
2392 
     <dd>Recursive PARTitioning and regression trees.  <em>Recommended</em>.
2388 
<br><dt><strong>rpvm</strong>
 2393 
<br><dt><strong>rpvm</strong>
2389 
     <dd>R interface to PVM (Parallel Virtual Machine).  Provides interface to
 2394 
     <dd>R interface to PVM (Parallel Virtual Machine).  Provides interface to
2390 
PVM APIs, and examples and documentation for its use.
2395 
PVM APIs, and examples and documentation for its use.
2391 
<br><dt><strong>rqmcmb2</strong>
 2396 
<br><dt><strong>rqmcmb2</strong>
2392 
     <dd>Markov chain marginal bootstrap for quantile regression.
2397 
     <dd>Markov chain marginal bootstrap for quantile regression.
2393 
<br><dt><strong>rsprng</strong>
 2398 
<br><dt><strong>rsprng</strong>
2394 
     <dd>Provides interface to SPRNG (Scalable Parallel Random Number Generators)
 2399 
     <dd>Provides interface to SPRNG (Scalable Parallel Random Number Generators)
2395 
APIs, and examples and documentation for its use.
2400 
APIs, and examples and documentation for its use.
2396 
<br><dt><strong>sampfling</strong>
 2401 
<br><dt><strong>sampfling</strong>
2397 
     <dd>Implements a modified version of the Sampford sampling algorithm.  Given
 2402 
     <dd>Implements a modified version of the Sampford sampling algorithm.  Given
2398 
a quantity assigned to each unit in the population, samples are drawn
 2403 
a quantity assigned to each unit in the population, samples are drawn
2399 
with probability proportional to te product of the quantities of the
 2404 
with probability proportional to te product of the quantities of the
2400 
units included in the sample.
2405 
units included in the sample.
2401 
<br><dt><strong>sca</strong>
 2406 
<br><dt><strong>sca</strong>
2402 
     <dd>Simple Component Analysis.
2407 
     <dd>Simple Component Analysis.
2403 
<br><dt><strong>scatterplot3d</strong>
 2408 
<br><dt><strong>scatterplot3d</strong>
2404 
     <dd>Plots a three dimensional (3D) point cloud perspectively.
2409 
     <dd>Plots a three dimensional (3D) point cloud perspectively.
2405 
<br><dt><strong>seacarb</strong>
 2410 
<br><dt><strong>seacarb</strong>
2406 
     <dd>Calculates parameters of the seawater carbonate system.
2411 
     <dd>Calculates parameters of the seawater carbonate system.
2407 
<br><dt><strong>seao</strong>
 2412 
<br><dt><strong>seao</strong>
2408 
     <dd>Simple Evolutionary Algorithm Optimization.
2413 
     <dd>Simple Evolutionary Algorithm Optimization.
2409 
<br><dt><strong>seao.gui</strong>
 2414 
<br><dt><strong>seao.gui</strong>
2410 
     <dd>Simple Evolutionary Algorithm Optimization: graphical user interface.
2415 
     <dd>Simple Evolutionary Algorithm Optimization: graphical user interface.
2411 
<br><dt><strong>segmented</strong>
 2416 
<br><dt><strong>segmented</strong>
2412 
     <dd>Functions to estimate break-points of segmented relationships in
 2417 
     <dd>Functions to estimate break-points of segmented relationships in
2413 
regression models (GLMs).
2418 
regression models (GLMs).
2414 
<br><dt><strong>sem</strong>
 2419 
<br><dt><strong>sem</strong>
2415 
     <dd>Functions for fitting general linear Structural Equation Models (with
 2420 
     <dd>Functions for fitting general linear Structural Equation Models (with
2416 
observed and unobserved variables) by the method of maximum likelihood
 2421 
observed and unobserved variables) by the method of maximum likelihood
2417 
using the RAM approach.
2422 
using the RAM approach.
2418 
<br><dt><strong>serialize</strong>
 2423 
<br><dt><strong>serialize</strong>
2419 
     <dd>Simple interfce for serializing to connections.
2424 
     <dd>Simple interfce for serializing to connections.
2420 
<br><dt><strong>session</strong>
 2425 
<br><dt><strong>session</strong>
2421 
     <dd>Functions for interacting with, saving and restoring R sessions.
2426 
     <dd>Functions for interacting with, saving and restoring R sessions.
2422 
<br><dt><strong>sfsmisc</strong>
 2427 
<br><dt><strong>sfsmisc</strong>
2423 
     <dd>Utilities from Seminar fuer Statistik ETH Zurich.
2428 
     <dd>Utilities from Seminar fuer Statistik ETH Zurich.
2424 
<br><dt><strong>sgeostat</strong>
 2429 
<br><dt><strong>sgeostat</strong>
2425 
     <dd>An object-oriented framework for geostatistical modeling.
2430 
     <dd>An object-oriented framework for geostatistical modeling.
2426 
<br><dt><strong>shapefiles</strong>
 2431 
<br><dt><strong>shapefiles</strong>
2427 
     <dd>Functions to read and write ESRI shapefiles.
2432 
     <dd>Functions to read and write ESRI shapefiles.
2428 
<br><dt><strong>shapes</strong>
 2433 
<br><dt><strong>shapes</strong>
2429 
     <dd>Routines for the statistical analysis of shapes, including procrustes
 2434 
     <dd>Routines for the statistical analysis of shapes, including procrustes
2430 
analysis, displaying shapes and principal components, testing for mean
 2435 
analysis, displaying shapes and principal components, testing for mean
2431 
shape difference, thin-plate spline transformation grids and edge
 2436 
shape difference, thin-plate spline transformation grids and edge
2432 
superimposition methods.
2437 
superimposition methods.
2433 
<br><dt><strong>simpleboot</strong>
 2438 
<br><dt><strong>simpleboot</strong>
2434 
     <dd>Simple bootstrap routines.
2439 
     <dd>Simple bootstrap routines.
2435 
<br><dt><strong>sm</strong>
 2440 
<br><dt><strong>sm</strong>
2436 
     <dd>Software linked to the book "Applied Smoothing Techniques for Data
 2441 
     <dd>Software linked to the book "Applied Smoothing Techniques for Data
2437 
Analysis:  The Kernel Approach with <small>S-PLUS</small> Illustrations" by
 2442 
Analysis:  The Kernel Approach with <small>S-PLUS</small> Illustrations" by
2438 
A. W. Bowman and A. Azzalini (1997), Oxford University Press.
2443 
A. W. Bowman and A. Azzalini (1997), Oxford University Press.
2439 
<br><dt><strong>sma</strong>
 2444 
<br><dt><strong>sma</strong>
2440 
     <dd>Functions for exploratory (statistical) microarray analysis.
2445 
     <dd>Functions for exploratory (statistical) microarray analysis.
2441 
<br><dt><strong>smoothSurv</strong>
 2446 
<br><dt><strong>smoothSurv</strong>
2442 
     <dd>Survival regression with smoothed error distribution.
2447 
     <dd>Survival regression with smoothed error distribution.
2443 
<br><dt><strong>sn</strong>
 2448 
<br><dt><strong>sn</strong>
2444 
     <dd>Functions for manipulating skew-normal probability distributions and for
 2449 
     <dd>Functions for manipulating skew-normal probability distributions and for
2445 
fitting them to data, in the scalar and the multivariate case.
2450 
fitting them to data, in the scalar and the multivariate case.
2446 
<br><dt><strong>sna</strong>
 2451 
<br><dt><strong>sna</strong>
2447 
     <dd>A range of tools for social network analysis, including node and
 2452 
     <dd>A range of tools for social network analysis, including node and
2448 
graph-level indices, structural distance and covariance methods,
 2453 
graph-level indices, structural distance and covariance methods,
2449 
structural equivalence detection, p* modeling, and network
 2454 
structural equivalence detection, p* modeling, and network
2450 
visualization.
2455 
visualization.
2451 
<br><dt><strong>snow</strong>
 2456 
<br><dt><strong>snow</strong>
2452 
     <dd>Simple Network of Workstations: support for simple parallel computing in
 2457 
     <dd>Simple Network of Workstations: support for simple parallel computing in
2453 
R.
2458 
R.
2454 
<br><dt><strong>som</strong>
 2459 
<br><dt><strong>som</strong>
2455 
     <dd>Self-Organizing Maps (with application in gene clustering).
2460 
     <dd>Self-Organizing Maps (with application in gene clustering).
2456 
<br><dt><strong>sound</strong>
 2461 
<br><dt><strong>sound</strong>
2457 
     <dd>A sound interface for R: Basic functions for dealing with <code>.wav</code>
 2462 
     <dd>A sound interface for R: Basic functions for dealing with <code>.wav</code>
2458 
files and sound samples.
2463 
files and sound samples.
2459 
<br><dt><strong>spatial</strong>
 2464 
<br><dt><strong>spatial</strong>
2460 
     <dd>Functions for kriging and point pattern analysis from "Modern Applied
 2465 
     <dd>Functions for kriging and point pattern analysis from "Modern Applied
2461 
Statistics with S" by W. Venables and B. Ripley.  Contained in the
 2466 
Statistics with S" by W. Venables and B. Ripley.  Contained in the
2462 
<code>VR</code> bundle.  <em>Recommended</em>.
2467 
<code>VR</code> bundle.  <em>Recommended</em>.
2463 
<br><dt><strong>spatstat</strong>
 2468 
<br><dt><strong>spatstat</strong>
2464 
     <dd>Data analysis and modelling of two-dimensional point patterns, including
 2469 
     <dd>Data analysis and modelling of two-dimensional point patterns, including
2465 
multitype points and spatial covariates.
2470 
multitype points and spatial covariates.
2466 
<br><dt><strong>spdep</strong>
 2471 
<br><dt><strong>spdep</strong>
2467 
     <dd>A collection of functions to create spatial weights matrix objects from
 2472 
     <dd>A collection of functions to create spatial weights matrix objects from
2468 
polygon contiguities, from point patterns by distance and tesselations,
 2473 
polygon contiguities, from point patterns by distance and tesselations,
2469 
for summarising these objects, and for permitting their use in spatial
 2474 
for summarising these objects, and for permitting their use in spatial
2470 
data analysis; a collection of tests for spatial autocorrelation,
 2475 
data analysis; a collection of tests for spatial autocorrelation,
2471 
including global Moran's I and Geary's C, local Moran's I, saddlepoint
 2476 
including global Moran's I and Geary's C, local Moran's I, saddlepoint
2472 
approximations for global and local Moran's I; and functions for
 2477 
approximations for global and local Moran's I; and functions for
2473 
estimating spatial simultaneous autoregressive (SAR) models.  (Was
 2478 
estimating spatial simultaneous autoregressive (SAR) models.  (Was
2474 
formerly the three packages: <strong>spweights</strong>, <strong>sptests</strong>, and
 2479 
formerly the three packages: <strong>spweights</strong>, <strong>sptests</strong>, and
2475 
<strong>spsarlm</strong>.)
2480 
<strong>spsarlm</strong>.)
2476 
<br><dt><strong>splancs</strong>
 2481 
<br><dt><strong>splancs</strong>
2477 
     <dd>Spatial and space-time point pattern analysis functions.
2482 
     <dd>Spatial and space-time point pattern analysis functions.
2478 
<br><dt><strong>statmod</strong>
 2483 
<br><dt><strong>statmod</strong>
2479 
     <dd>Miscellaneous biostatistical modelling functions.
2484 
     <dd>Miscellaneous biostatistical modelling functions.
2480 
<br><dt><strong>strucchange</strong>
 2485 
<br><dt><strong>strucchange</strong>
2481 
     <dd>Various tests on structural change in linear regression models.
2486 
     <dd>Various tests on structural change in linear regression models.
2482 
<br><dt><strong>subselect</strong>
 2487 
<br><dt><strong>subselect</strong>
2483 
     <dd>A collection of functions which assess the quality of variable subsets
 2488 
     <dd>A collection of functions which assess the quality of variable subsets
2484 
as surrogates for a full data set, and search for subsets which are
 2489 
as surrogates for a full data set, and search for subsets which are
2485 
optimal under various criteria.
2490 
optimal under various criteria.
2486 
<br><dt><strong>supclust</strong>
 2491 
<br><dt><strong>supclust</strong>
2487 
     <dd>Methodology for supervised grouping of predictor variables.
2492 
     <dd>Methodology for supervised grouping of predictor variables.
2488 
<br><dt><strong>survey</strong>
 2493 
<br><dt><strong>survey</strong>
2489 
     <dd>Summary statistics, generalized linear models, and general maximum
 2494 
     <dd>Summary statistics, generalized linear models, and general maximum
2490 
likelihood estimation for stratified, cluster-sampled, unequally
 2495 
likelihood estimation for stratified, cluster-sampled, unequally
2491 
weighted survey samples.
2496 
weighted survey samples.
2492 
<br><dt><strong>survival</strong>
 2497 
<br><dt><strong>survival</strong>
2493 
     <dd>Functions for survival analysis, including penalised likelihood.
2498 
     <dd>Functions for survival analysis, including penalised likelihood.
2494 
<em>Recommended</em>.
2499 
<em>Recommended</em>.
2495 
<br><dt><strong>survrec</strong>
 2500 
<br><dt><strong>survrec</strong>
2496 
     <dd>Survival analysis for recurrent event data.
2501 
     <dd>Survival analysis for recurrent event data.
2497 
<br><dt><strong>systemfit</strong>
 2502 
<br><dt><strong>systemfit</strong>
2498 
     <dd>Contains functions for fitting simultaneous systems of equations using
 2503 
     <dd>Contains functions for fitting simultaneous systems of equations using
2499 
Ordinary Least Sqaures (OLS), Two-Stage Least Squares (2SLS), and
 2504 
Ordinary Least Sqaures (OLS), Two-Stage Least Squares (2SLS), and
2500 
Three-Stage Least Squares (3SLS).
2505 
Three-Stage Least Squares (3SLS).
2501 
<br><dt><strong>tapiR</strong>
 2506 
<br><dt><strong>tapiR</strong>
2502 
     <dd>Tools for accessing (UK) parliamentary information in R.
2507 
     <dd>Tools for accessing (UK) parliamentary information in R.
2503 
<br><dt><strong>tensor</strong>
 2508 
<br><dt><strong>tensor</strong>
2504 
     <dd>Tensor product of arrays.
2509 
     <dd>Tensor product of arrays.
2505 
<br><dt><strong>tkrplot</strong>
 2510 
<br><dt><strong>tkrplot</strong>
2506 
     <dd>Simple mechanism for placing R graphics in a Tk widget.
2511 
     <dd>Simple mechanism for placing R graphics in a Tk widget.
2507 
<br><dt><strong>tree</strong>
 2512 
<br><dt><strong>tree</strong>
2508 
     <dd>Classification and regression trees.
2513 
     <dd>Classification and regression trees.
2509 
<br><dt><strong>tripack</strong>
 2514 
<br><dt><strong>tripack</strong>
2510 
     <dd>A constrained two-dimensional Delaunay triangulation package.
2515 
     <dd>A constrained two-dimensional Delaunay triangulation package.
2511 
<br><dt><strong>tseries</strong>
 2516 
<br><dt><strong>tseries</strong>
2512 
     <dd>Package for time series analysis with emphasis on non-linear modelling.
2517 
     <dd>Package for time series analysis with emphasis on non-linear modelling.
2513 
<br><dt><strong>twostage</strong>
 2518 
<br><dt><strong>twostage</strong>
2514 
     <dd>Functions for optimal design of two-stage-studies using the Mean Score
 2519 
     <dd>Functions for optimal design of two-stage-studies using the Mean Score
2515 
method.
2520 
method.
2516 
<br><dt><strong>udunits</strong>
 2521 
<br><dt><strong>udunits</strong>
2517 
     <dd>Interface to Unidata's routines to convert units.
2522 
     <dd>Interface to Unidata's routines to convert units.
2518 
<br><dt><strong>vardiag</strong>
 2523 
<br><dt><strong>vardiag</strong>
2519 
     <dd>Interactive variogram diagnostics.
2524 
     <dd>Interactive variogram diagnostics.
2520 
<br><dt><strong>vcd</strong>
 2525 
<br><dt><strong>vcd</strong>
2521 
     <dd>Functions and data sets based on the book "Visualizing Categorical
 2526 
     <dd>Functions and data sets based on the book "Visualizing Categorical
2522 
Data" by Michael Friendly.
2527 
Data" by Michael Friendly.
2523 
<br><dt><strong>vegan</strong>
 2528 
<br><dt><strong>vegan</strong>
2524 
     <dd>Various help functions for vegetation scientists and community
 2529 
     <dd>Various help functions for vegetation scientists and community
2525 
ecologists.
2530 
ecologists.
2526 
<br><dt><strong>waveslim</strong>
 2531 
<br><dt><strong>waveslim</strong>
2527 
     <dd>Basic wavelet routines for time series analysis.
2532 
     <dd>Basic wavelet routines for time series analysis.
2528 
<br><dt><strong>wavethresh</strong>
 2533 
<br><dt><strong>wavethresh</strong>
2529 
     <dd>Software to perform 1-d and 2-d wavelet statistics and transforms.
2534 
     <dd>Software to perform 1-d and 2-d wavelet statistics and transforms.
2530 
<br><dt><strong>wle</strong>
 2535 
<br><dt><strong>wle</strong>
2531 
     <dd>Robust statistical inference via a weighted likelihood approach.
2536 
     <dd>Robust statistical inference via a weighted likelihood approach.
2532 
<br><dt><strong>xgobi</strong>
 2537 
<br><dt><strong>xgobi</strong>
2533 
     <dd>Interface to the XGobi and XGvis programs for graphical data analysis.
2538 
     <dd>Interface to the XGobi and XGvis programs for graphical data analysis.
2534 
<br><dt><strong>xtable</strong>
 2539 
<br><dt><strong>xtable</strong>
2535 
     <dd>Export data to LaTeX and <small>HTML</small> tables.
2540 
     <dd>Export data to LaTeX and <small>HTML</small> tables.
2536 
</dl>
 2541 
</dl>
2537 
 
 2542 
 
2538 
<p>See <small>CRAN</small> <code>src/contrib/PACKAGES</code> for more information.
 2543 
<p>See <small>CRAN</small> <code>src/contrib/PACKAGES</code> for more information.
2539 
 
 2544 
 
2540 
   <p>There is also a <small>CRAN</small> <code>src/contrib/Devel</code> directory which
 2545 
   <p>There is also a <small>CRAN</small> <code>src/contrib/Devel</code> directory which
2541 
contains packages still "under development" or depending on features
 2546 
contains packages still "under development" or depending on features
2542 
only present in the current development versions of R.  Volunteers are
 2547 
only present in the current development versions of R.  Volunteers are
2543 
invited to give these a try, of course.  This area of <small>CRAN</small> currently
 2548 
invited to give these a try, of course.  This area of <small>CRAN</small> currently
2544 
contains
 2549 
contains
2545 
 
 2550 
 
2546 
     <dl>
 2551 
     <dl>
2547 
<dt><strong>Dopt</strong>
 2552 
<dt><strong>Dopt</strong>
2548 
     <dd>Finding D-optimal experimental designs.
 2553 
     <dd>Finding D-optimal experimental designs.
2549 
 
 2554 
 
2550 
     <br><dt><strong>GLMMGibbs</strong>
 2555 
     <br><dt><strong>GLMMGibbs</strong>
2551 
     <dd>Generalised Linear Mixed Models by Gibbs sampling.
 2556 
     <dd>Generalised Linear Mixed Models by Gibbs sampling.
2552 
 
 2557 
 
2553 
     <br><dt><strong>RPgSQL</strong>
 2558 
     <br><dt><strong>RPgSQL</strong>
2554 
     <dd>Provides methods for accessing data stored in PostgreSQL tables.
 2559 
     <dd>Provides methods for accessing data stored in PostgreSQL tables.
2555 
 
 2560 
 
2556 
     <br><dt><strong>Rmpi</strong>
 2561 
     <br><dt><strong>Rmpi</strong>
2557 
     <dd>An interface (wrapper) to MPI (Message-Passing Interface) APIs.  It also
 2562 
     <dd>An interface (wrapper) to MPI (Message-Passing Interface) APIs.  It also
2558 
provides interactive R slave functionalities to make MPI programming
 2563 
provides interactive R slave functionalities to make MPI programming
2559 
easier in R than in C(++) or FORTRAN.
 2564 
easier in R than in C(++) or FORTRAN.
2560 
 
 2565 
 
2561 
     <br><dt><strong>dseplus</strong>
 2566 
     <br><dt><strong>dseplus</strong>
2562 
     <dd>Extensions to <strong>dse</strong>, the Dynamic Systems Estimation multivariate
 2567 
     <dd>Extensions to <strong>dse</strong>, the Dynamic Systems Estimation multivariate
2563 
time series package.  Contains PADI, juice and monitoring extensions.
 2568 
time series package.  Contains PADI, juice and monitoring extensions.
2564 
 
 2569 
 
2565 
     <br><dt><strong>ensemble</strong>
 2570 
     <br><dt><strong>ensemble</strong>
2566 
     <dd>Ensembles of tree classifiers.
 2571 
     <dd>Ensembles of tree classifiers.
2567 
 
 2572 
 
2568 
     <br><dt><strong>runStat</strong>
 2573 
     <br><dt><strong>runStat</strong>
2569 
     <dd>Running median and mean.
 2574 
     <dd>Running median and mean.
2570 
 
 2575 
 
2571 
     <br><dt><strong>write.snns</strong>
 2576 
     <br><dt><strong>write.snns</strong>
2572 
     <dd>Function for writing a <small>SNNS</small> pattern file from a data frame or
 2577 
     <dd>Function for writing a <small>SNNS</small> pattern file from a data frame or
2573 
matrix.
2578 
matrix.
2574 
</dl>
 2579 
</dl>
2575 
 
 2580 
 
2576 
<div class="node">
 2581 
<div class="node">
2577 
<p><hr>
 2582 
<p><hr>
2578 
Node:&nbsp;<a name="Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
 2583 
Node:&nbsp;<a name="Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
2579 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
 2584 
Next:&nbsp;<a rel="next" accesskey="n" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
2580 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
 2585 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20CRAN">Add-on packages from CRAN</a>,
2581 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
 2586 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
2582 
<br>
 2587 
<br>
2583 
</div>
 2588 
</div>
2584 
 
 2589 
 
2585 
<h3 class="subsection">5.1.3 Add-on packages from Omegahat</h4>
 2590 
<h3 class="subsection">5.1.3 Add-on packages from Omegahat</h4>
2586 
 
 2591 
 
2587 
<p>The <code>src/contrib/Omegahat</code> Directory of a <small>CRAN</small> site contains yet
 2592 
<p>The <code>src/contrib/Omegahat</code> Directory of a <small>CRAN</small> site contains yet
2588 
unreleased packages from the <a href="http://www.omegahat.org/">Omegahat Project for Statistical Computing</a>.  Currently, there are
 2593 
unreleased packages from the <a href="http://www.omegahat.org/">Omegahat Project for Statistical Computing</a>.  Currently, there are
2589 
 
 2594 
 
2590 
     <dl>
 2595 
     <dl>
2591 
<dt><strong>CORBA</strong>
 2596 
<dt><strong>CORBA</strong>
2592 
     <dd>Dynamic CORBA client/server facilities for R.  Connects to other
 2597 
     <dd>Dynamic CORBA client/server facilities for R.  Connects to other
2593 
CORBA-aware applications developed in arbitrary languages, on different
 2598 
CORBA-aware applications developed in arbitrary languages, on different
2594 
machines and allows R functionality to be exported in the same way to
 2599 
machines and allows R functionality to be exported in the same way to
2595 
other applications.
2600 
other applications.
2596 
<br><dt><strong>OOP</strong>
 2601 
<br><dt><strong>OOP</strong>
2597 
     <dd>OOP style classes and methods for R and <small>S-PLUS</small>.  Object references and
 2602 
     <dd>OOP style classes and methods for R and <small>S-PLUS</small>.  Object references and
2598 
class-based method definition are supported in the style of languages
 2603 
class-based method definition are supported in the style of languages
2599 
such as Java and C++.
2604 
such as Java and C++.
2600 
<br><dt><strong>REmbeddedPostgres</strong>
 2605 
<br><dt><strong>REmbeddedPostgres</strong>
2601 
     <dd>Allows R functions and objects to be used to implement SQL functions --
 2606 
     <dd>Allows R functions and objects to be used to implement SQL functions --
2602 
per-record, aggregate and trigger functions.
2607 
per-record, aggregate and trigger functions.
2603 
<br><dt><strong>REventLoop</strong>
 2608 
<br><dt><strong>REventLoop</strong>
2604 
     <dd>An abstract event loop mechanism that is toolkit independent and can be
 2609 
     <dd>An abstract event loop mechanism that is toolkit independent and can be
2605 
used to to replace the R event loop.
2610 
used to to replace the R event loop.
2606 
<br><dt><strong>RGdkPixbuf</strong>
 2611 
<br><dt><strong>RGdkPixbuf</strong>
2607 
     <dd>S language functions to access the facilities in the GdkPixbuf library
 2612 
     <dd>S language functions to access the facilities in the GdkPixbuf library
2608 
for manipulating images.
2613 
for manipulating images.
2609 
<br><dt><strong>RGnumeric</strong>
 2614 
<br><dt><strong>RGnumeric</strong>
2610 
     <dd>A plugin for the Gnumeric spreadsheet that allows R functions to be
 2615 
     <dd>A plugin for the Gnumeric spreadsheet that allows R functions to be
2611 
called from cells within the sheet, automatic recalculation, etc.
2616 
called from cells within the sheet, automatic recalculation, etc.
2612 
<br><dt><strong>RGtk</strong>
 2617 
<br><dt><strong>RGtk</strong>
2613 
     <dd>Facilities in the S language for programming graphical interfaces using
 2618 
     <dd>Facilities in the S language for programming graphical interfaces using
2614 
Gtk, the Gnome GUI toolkit.
2619 
Gtk, the Gnome GUI toolkit.
2615 
<br><dt><strong>RGtkBindingGenerator</strong>
 2620 
<br><dt><strong>RGtkBindingGenerator</strong>
2616 
     <dd>A meta-package which generates C and R code to provide bindings to a
 2621 
     <dd>A meta-package which generates C and R code to provide bindings to a
2617 
Gtk-based library.
2622 
Gtk-based library.
2618 
<br><dt><strong>RGtkExtra</strong>
 2623 
<br><dt><strong>RGtkExtra</strong>
2619 
     <dd>A collection of S functions that provide an interface to the widgets in
 2624 
     <dd>A collection of S functions that provide an interface to the widgets in
2620 
the gtk+extra library such as the GtkSheet data-grid display, icon list,
 2625 
the gtk+extra library such as the GtkSheet data-grid display, icon list,
2621 
file list and directory tree.
2626 
file list and directory tree.
2622 
<br><dt><strong>RGtkGlade</strong>
 2627 
<br><dt><strong>RGtkGlade</strong>
2623 
     <dd>S language bindings providing an interface to Glade, the interactive
 2628 
     <dd>S language bindings providing an interface to Glade, the interactive
2624 
Gnome GUI creator.
2629 
Gnome GUI creator.
2625 
<br><dt><strong>RGtkHTML</strong>
 2630 
<br><dt><strong>RGtkHTML</strong>
2626 
     <dd>A collection of S functions that provide an interface to creating and
 2631 
     <dd>A collection of S functions that provide an interface to creating and
2627 
controlling an <small>HTML</small> widget which can be used to display <small>HTML</small>
 2632 
controlling an <small>HTML</small> widget which can be used to display <small>HTML</small>
2628 
documents from files or content generated dynamically in S.
2633 
documents from files or content generated dynamically in S.
2629 
<br><dt><strong>RGtkViewers</strong>
 2634 
<br><dt><strong>RGtkViewers</strong>
2630 
     <dd>A collection of tools for viewing different S objects, databases, class
 2635 
     <dd>A collection of tools for viewing different S objects, databases, class
2631 
and widget hierarchies, S source file contents, etc.
2636 
and widget hierarchies, S source file contents, etc.
2632 
<br><dt><strong>RJavaDevice</strong>
 2637 
<br><dt><strong>RJavaDevice</strong>
2633 
     <dd>A graphics device for R that uses Java components and graphics.
2638 
     <dd>A graphics device for R that uses Java components and graphics.
2634 
<small>API</small>s.
2639 
<small>API</small>s.
2635 
<br><dt><strong>RObjectTables</strong>
 2640 
<br><dt><strong>RObjectTables</strong>
2636 
     <dd>The C and S code allows one to define R objects to be used as elements
 2641 
     <dd>The C and S code allows one to define R objects to be used as elements
2637 
of the search path with their own semantics and facilities for reading
 2642 
of the search path with their own semantics and facilities for reading
2638 
and writing variables.  The objects implement a simple interface via R
 2643 
and writing variables.  The objects implement a simple interface via R
2639 
functions (either methods or closures) and can access external data,
 2644 
functions (either methods or closures) and can access external data,
2640 
e.g., in other applications, languages, formats, <small class="dots">...</small>
 2645 
e.g., in other applications, languages, formats, <small class="dots">...</small>
2641 
<br><dt><strong>RSMethods</strong>
 2646 
<br><dt><strong>RSMethods</strong>
2642 
     <dd>An implementation of S version 4 methods and classes for R, consistent
 2647 
     <dd>An implementation of S version 4 methods and classes for R, consistent
2643 
with the basic material in "Programming with Data" by John
 2648 
with the basic material in "Programming with Data" by John
2644 
M. Chambers, 1998, Springer NY.
2649 
M. Chambers, 1998, Springer NY.
2645 
<br><dt><strong>RSPerl</strong>
 2650 
<br><dt><strong>RSPerl</strong>
2646 
     <dd>An interface from R to an embedded, persistent Perl interpreter,
 2651 
     <dd>An interface from R to an embedded, persistent Perl interpreter,
2647 
allowing one to call arbitrary Perl subroutines, classes and methods.
2652 
allowing one to call arbitrary Perl subroutines, classes and methods.
2648 
<br><dt><strong>RSPython</strong>
 2653 
<br><dt><strong>RSPython</strong>
2649 
     <dd>Allows Python programs to invoke S functions, methods, etc., and S code
 2654 
     <dd>Allows Python programs to invoke S functions, methods, etc., and S code
2650 
to call Python functionality.
2655 
to call Python functionality.
2651 
<br><dt><strong>RXLisp</strong>
 2656 
<br><dt><strong>RXLisp</strong>
2652 
     <dd>An interface to call XLisp-Stat functions from within R.
2657 
     <dd>An interface to call XLisp-Stat functions from within R.
2653 
<br><dt><strong>SASXML</strong>
 2658 
<br><dt><strong>SASXML</strong>
2654 
     <dd>Example for reading <small>XML</small> files in SAS 8.2 manner.
2659 
     <dd>Example for reading <small>XML</small> files in SAS 8.2 manner.
2655 
<br><dt><strong>SJava</strong>
 2660 
<br><dt><strong>SJava</strong>
2656 
     <dd>An interface from R to Java to create and call Java objects and
 2661 
     <dd>An interface from R to Java to create and call Java objects and
2657 
methods.
2662 
methods.
2658 
<br><dt><strong>SLanguage</strong>
 2663 
<br><dt><strong>SLanguage</strong>
2659 
     <dd>Functions and C support utilities to support S language programming
 2664 
     <dd>Functions and C support utilities to support S language programming
2660 
that can work in both R and <small>S-PLUS</small>.
2665 
that can work in both R and <small>S-PLUS</small>.
2661 
<br><dt><strong>SNetscape</strong>
 2666 
<br><dt><strong>SNetscape</strong>
2662 
     <dd>Plugin for Netscape and JavaScript.
2667 
     <dd>Plugin for Netscape and JavaScript.
2663 
<br><dt><strong>SWinRegistry</strong>
 2668 
<br><dt><strong>SWinRegistry</strong>
2664 
     <dd>Provides access from within R to read and write the Windows registry.
2669 
     <dd>Provides access from within R to read and write the Windows registry.
2665 
<br><dt><strong>SWinTypeLibs</strong>
 2670 
<br><dt><strong>SWinTypeLibs</strong>
2666 
     <dd>Provides ways to extract type information from type libraries and/or
 2671 
     <dd>Provides ways to extract type information from type libraries and/or
2667 
DCOM objects that describes the methods, properties, etc. of an
 2672 
DCOM objects that describes the methods, properties, etc. of an
2668 
interface.
2673 
interface.
2669 
<br><dt><strong>SXalan</strong>
 2674 
<br><dt><strong>SXalan</strong>
2670 
     <dd>Process <small>XML</small> documents using <small>XSL</small> functions implemented in R and
 2675 
     <dd>Process <small>XML</small> documents using <small>XSL</small> functions implemented in R and
2671 
dynamically substituting output from R.
2676 
dynamically substituting output from R.
2672 
<br><dt><strong>Slcc</strong>
 2677 
<br><dt><strong>Slcc</strong>
2673 
     <dd>Parses C source code, allowing one to analyze and automatically generate
 2678 
     <dd>Parses C source code, allowing one to analyze and automatically generate
2674 
interfaces from S to that code, including the table of S-accessible
 2679 
interfaces from S to that code, including the table of S-accessible
2675 
native symbols, parameter count and type information, S constructors
 2680 
native symbols, parameter count and type information, S constructors
2676 
from C objects, call graphs, etc.
2681 
from C objects, call graphs, etc.
2677 
<br><dt><strong>Sxslt</strong>
 2682 
<br><dt><strong>Sxslt</strong>
2678 
     <dd>An extension module for libxslt, the <small>XML</small>-<small>XSL</small> document translator,
 2683 
     <dd>An extension module for libxslt, the <small>XML</small>-<small>XSL</small> document translator,
2679 
that allows <small>XSL</small> functions to be implemented via R functions.
2684 
that allows <small>XSL</small> functions to be implemented via R functions.
2680 
</dl>
 2685 
</dl>
2681 
 
 2686 
 
2682 
<div class="node">
 2687 
<div class="node">
2683 
<p><hr>
 2688 
<p><hr>
2684 
Node:&nbsp;<a name="Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
 2689 
Node:&nbsp;<a name="Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
2685 
Next:&nbsp;<a rel="next" accesskey="n" href="#Other%20add-on%20packages">Other add-on packages</a>,
 2690 
Next:&nbsp;<a rel="next" accesskey="n" href="#Other%20add-on%20packages">Other add-on packages</a>,
2686 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
 2691 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20Omegahat">Add-on packages from Omegahat</a>,
2687 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
 2692 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
2688 
<br>
 2693 
<br>
2689 
</div>
 2694 
</div>
2690 
 
 2695 
 
2691 
<h3 class="subsection">5.1.4 Add-on packages from BioConductor</h4>
 2696 
<h3 class="subsection">5.1.4 Add-on packages from BioConductor</h4>
2692 
 
 2697 
 
2693 
<p>The <a href="http://www.bioconductor.org">Bioconductor Project</a> produces an
 2698 
<p>The <a href="http://www.bioconductor.org">Bioconductor Project</a> produces an
2694 
open source software framework that will assist biologists and
 2699 
open source software framework that will assist biologists and
2695 
statisticians working in bioinformatics, with primary emphasis on
 2700 
statisticians working in bioinformatics, with primary emphasis on
2696 
inference using DNA microarrays.  The following R packages are contained
 2701 
inference using DNA microarrays.  The following R packages are contained
2697 
in the current release of BioConductor, with more packages under
 2702 
in the current release of BioConductor, with more packages under
2698 
development.
 2703 
development.
2699 
 
 2704 
 
2700 
     <dl>
 2705 
     <dl>
2701 
<dt><strong>AnnBuilder</strong>
 2706 
<dt><strong>AnnBuilder</strong>
2702 
     <dd>Assemble and process genomic annotation data, from databases such as
 2707 
     <dd>Assemble and process genomic annotation data, from databases such as
2703 
GenBank, the Gene Ontology Consortium, LocusLink, UniGene, the UCSC
 2708 
GenBank, the Gene Ontology Consortium, LocusLink, UniGene, the UCSC
2704 
Human Genome Project.
2709 
Human Genome Project.
2705 
<br><dt><strong>Biobase</strong>
 2710 
<br><dt><strong>Biobase</strong>
2706 
     <dd>Object-oriented representation and manipulation of genomic data (S4
 2711 
     <dd>Object-oriented representation and manipulation of genomic data (S4
2707 
class structure).
2712 
class structure).
2708 
<br><dt><strong>DynDoc</strong>
 2713 
<br><dt><strong>DynDoc</strong>
2709 
     <dd>Functionality to create and interact with dynamic documents, vignettes,
 2714 
     <dd>Functionality to create and interact with dynamic documents, vignettes,
2710 
and other navigable documents.
2715 
and other navigable documents.
2711 
<br><dt><strong>MAGEML</strong>
 2716 
<br><dt><strong>MAGEML</strong>
2712 
     <dd>Functionality to handle MAGEML documents.
2717 
     <dd>Functionality to handle MAGEML documents.
2713 
<br><dt><strong>RBGL</strong>
 2718 
<br><dt><strong>RBGL</strong>
2714 
     <dd>An interface between the graph package and the Boost graph libraries,
 2719 
     <dd>An interface between the graph package and the Boost graph libraries,
2715 
allowing for fast manipulation of graph objects in R.
2720 
allowing for fast manipulation of graph objects in R.
2716 
<br><dt><strong>ROC</strong>
 2721 
<br><dt><strong>ROC</strong>
2717 
     <dd>Receiver Operating Characteristic (ROC) approach for identifying genes
 2722 
     <dd>Receiver Operating Characteristic (ROC) approach for identifying genes
2718 
that are differentially expressed in two types of samples.
2723 
that are differentially expressed in two types of samples.
2719 
<br><dt><strong>RdbiPgSQL</strong>
 2724 
<br><dt><strong>RdbiPgSQL</strong>
2720 
     <dd>Methods for accessing data stored in PostgreSQL tables.
2725 
     <dd>Methods for accessing data stored in PostgreSQL tables.
2721 
<br><dt><strong>Rgraphviz</strong>
 2726 
<br><dt><strong>Rgraphviz</strong>
2722 
     <dd>An interface with Graphviz for plotting graph objects in R.
2727 
     <dd>An interface with Graphviz for plotting graph objects in R.
2723 
<br><dt><strong>Ruuid</strong>
 2728 
<br><dt><strong>Ruuid</strong>
2724 
     <dd>Creates Universally Unique ID values (UUIDs) in R.
2729 
     <dd>Creates Universally Unique ID values (UUIDs) in R.
2725 
<br><dt><strong>SAGElyzer</strong>
 2730 
<br><dt><strong>SAGElyzer</strong>
2726 
     <dd>Locates genes based on SAGE tags.
2731 
     <dd>Locates genes based on SAGE tags.
2727 
<br><dt><strong>SNPtools</strong>
 2732 
<br><dt><strong>SNPtools</strong>
2728 
     <dd>Rudimentary structures for SNP data.
2733 
     <dd>Rudimentary structures for SNP data.
2729 
<br><dt><strong>affy</strong>
 2734 
<br><dt><strong>affy</strong>
2730 
     <dd>Methods for Affymetrix Oligonucleotide Arrays.
2735 
     <dd>Methods for Affymetrix Oligonucleotide Arrays.
2731 
<br><dt><strong>affyPLM</strong>
 2736 
<br><dt><strong>affyPLM</strong>
2732 
     <dd>For fitting Probe Level Models.
2737 
     <dd>For fitting Probe Level Models.
2733 
<br><dt><strong>affycomp</strong>
 2738 
<br><dt><strong>affycomp</strong>
2734 
     <dd>Graphics toolbox for assessment of Affymetrix expression measures.
2739 
     <dd>Graphics toolbox for assessment of Affymetrix expression measures.
2735 
<br><dt><strong>affydata</strong>
 2740 
<br><dt><strong>affydata</strong>
2736 
     <dd>Affymetrix data for demonstration purposes.
2741 
     <dd>Affymetrix data for demonstration purposes.
2737 
<br><dt><strong>annaffy</strong>
 2742 
<br><dt><strong>annaffy</strong>
2738 
     <dd>Functions for handling data from Bioconductor Affymetrix annotation data
 2743 
     <dd>Functions for handling data from Bioconductor Affymetrix annotation data
2739 
packages.
2744 
packages.
2740 
<br><dt><strong>annotate</strong>
 2745 
<br><dt><strong>annotate</strong>
2741 
     <dd>Associate experimental data in real time to biological metadata from web
 2746 
     <dd>Associate experimental data in real time to biological metadata from web
2742 
databases such as GenBank, LocusLink and PubMed.  Process and store
 2747 
databases such as GenBank, LocusLink and PubMed.  Process and store
2743 
query results.  Generate <small>HTML</small> reports of analyses.
2748 
query results.  Generate <small>HTML</small> reports of analyses.
2744 
<br><dt><strong>ctc</strong>
 2749 
<br><dt><strong>ctc</strong>
2745 
     <dd>Tools to export and import Tree and Cluster to other programs.
2750 
     <dd>Tools to export and import Tree and Cluster to other programs.
2746 
<br><dt><strong>daMA</strong>
 2751 
<br><dt><strong>daMA</strong>
2747 
     <dd>Functions for the efficient design of factorial two-color microarray
 2752 
     <dd>Functions for the efficient design of factorial two-color microarray
2748 
experiments and for the statistical analysis of factorial microarray
 2753 
experiments and for the statistical analysis of factorial microarray
2749 
data.
2754 
data.
2750 
<br><dt><strong>edd</strong>
 2755 
<br><dt><strong>edd</strong>
2751 
     <dd>Expression density diagnostics: graphical methods and pattern
 2756 
     <dd>Expression density diagnostics: graphical methods and pattern
2752 
recognition algorithms for distribution shape classification.
2757 
recognition algorithms for distribution shape classification.
2753 
<br><dt><strong>externalVector</strong>
 2758 
<br><dt><strong>externalVector</strong>
2754 
     <dd>Basic class definitions and generics for external pointer based vector
 2759 
     <dd>Basic class definitions and generics for external pointer based vector
2755 
objects for R.
2760 
objects for R.
2756 
<br><dt><strong>factDesign</strong>
 2761 
<br><dt><strong>factDesign</strong>
2757 
     <dd>A set of tools for analyzing data from factorial designed micraorray
 2762 
     <dd>A set of tools for analyzing data from factorial designed micraorray
2758 
experiments.  The functions can be used to evaluate appropriate tests of
 2763 
experiments.  The functions can be used to evaluate appropriate tests of
2759 
contrast and perform single outlier detection.
2764 
contrast and perform single outlier detection.
2760 
<br><dt><strong>gcrma</strong>
 2765 
<br><dt><strong>gcrma</strong>
2761 
     <dd>Background adjustment using sequence information.
2766 
     <dd>Background adjustment using sequence information.
2762 
<br><dt><strong>genefilter</strong>
 2767 
<br><dt><strong>genefilter</strong>
2763 
     <dd>Tools for sequentially filtering genes using a wide variety of filtering
 2768 
     <dd>Tools for sequentially filtering genes using a wide variety of filtering
2764 
functions.  Example of filters include: number of missing value,
 2769 
functions.  Example of filters include: number of missing value,
2765 
coefficient of variation of expression measures, ANOVA p-value,
 2770 
coefficient of variation of expression measures, ANOVA p-value,
2766 
Cox model p-values.  Sequential application of filtering
 2771 
Cox model p-values.  Sequential application of filtering
2767 
functions to genes.
2772 
functions to genes.
2768 
<br><dt><strong>geneplotter</strong>
 2773 
<br><dt><strong>geneplotter</strong>
2769 
     <dd>Graphical tools for genomic data, for example for plotting expression
 2774 
     <dd>Graphical tools for genomic data, for example for plotting expression
2770 
data along a chromosome or producing color images of expression data
 2775 
data along a chromosome or producing color images of expression data
2771 
matrices.
2776 
matrices.
2772 
<br><dt><strong>globaltest</strong>
 2777 
<br><dt><strong>globaltest</strong>
2773 
     <dd>Testing globally whether a group of genes is significantly related to
 2778 
     <dd>Testing globally whether a group of genes is significantly related to
2774 
some clinical variable of interest.
2779 
some clinical variable of interest.
2775 
<br><dt><strong>gpls</strong>
 2780 
<br><dt><strong>gpls</strong>
2776 
     <dd>Classification using generalized partial least squares for two-group and
 2781 
     <dd>Classification using generalized partial least squares for two-group and
2777 
multi-group classification.
2782 
multi-group classification.
2778 
<br><dt><strong>graph</strong>
 2783 
<br><dt><strong>graph</strong>
2779 
     <dd>Classes and tools for creating and manipulating graphs within R.
2784 
     <dd>Classes and tools for creating and manipulating graphs within R.
2780 
<br><dt><strong>hexbin</strong>
 2785 
<br><dt><strong>hexbin</strong>
2781 
     <dd>Binning functions, in particular hexagonal bins for graphing.
2786 
     <dd>Binning functions, in particular hexagonal bins for graphing.
2782 
<br><dt><strong>limma</strong>
 2787 
<br><dt><strong>limma</strong>
2783 
     <dd>Linear models for microarray data.
2788 
     <dd>Linear models for microarray data.
2784 
<br><dt><strong>makecdfenv</strong>
 2789 
<br><dt><strong>makecdfenv</strong>
2785 
     <dd>Two functions.  One reads a Affymetrix chip description file (CDF) and
 2790 
     <dd>Two functions.  One reads a Affymetrix chip description file (CDF) and
2786 
creates a hash table environment containing the location/probe set
 2791 
creates a hash table environment containing the location/probe set
2787 
membership mapping.  The other creates a package that automatically loads
 2792 
membership mapping.  The other creates a package that automatically loads
2788 
that environment.
2793 
that environment.
2789 
<br><dt><strong>marrayClasses</strong>
 2794 
<br><dt><strong>marrayClasses</strong>
2790 
     <dd>Class definitions for pre-normalized and normalized cDNA microarray
 2795 
     <dd>Class definitions for pre-normalized and normalized cDNA microarray
2791 
data.  Basic methods for accessing/replacing, printing, and subsetting.
2796 
data.  Basic methods for accessing/replacing, printing, and subsetting.
2792 
<br><dt><strong>marrayInput</strong>
 2797 
<br><dt><strong>marrayInput</strong>
2793 
     <dd>Functions for reading microarray data into R from different image
 2798 
     <dd>Functions for reading microarray data into R from different image
2794 
analysis output files, and probe and target description files.  Widgets
 2799 
analysis output files, and probe and target description files.  Widgets
2795 
are supplied to facilitate and automate data input and the creation of
 2800 
are supplied to facilitate and automate data input and the creation of
2796 
microarray specific R objects for storing these data.
2801 
microarray specific R objects for storing these data.
2797 
<br><dt><strong>marrayNorm</strong>
 2802 
<br><dt><strong>marrayNorm</strong>
2798 
     <dd>Functions for location and scale normalization procedures based on
 2803 
     <dd>Functions for location and scale normalization procedures based on
2799 
robust local regression.
2804 
robust local regression.
2800 
<br><dt><strong>marrayPlots</strong>
 2805 
<br><dt><strong>marrayPlots</strong>
2801 
     <dd>Functions for diagnostic plots for pre- and post-normalization cDNA
 2806 
     <dd>Functions for diagnostic plots for pre- and post-normalization cDNA
2802 
microarray intensity data: boxplots, scatter-plots, color images.
2807 
microarray intensity data: boxplots, scatter-plots, color images.
2803 
<br><dt><strong>marrayTools</strong>
 2808 
<br><dt><strong>marrayTools</strong>
2804 
     <dd>Miscellaneous functions used in the functional genomics core facility in
 2809 
     <dd>Miscellaneous functions used in the functional genomics core facility in
2805 
UCB and UCSF.
2810 
UCB and UCSF.
2806 
<br><dt><strong>matchprobes</strong>
 2811 
<br><dt><strong>matchprobes</strong>
2807 
     <dd>ools for sequence matching of probes on arrays.
2812 
     <dd>ools for sequence matching of probes on arrays.
2808 
<br><dt><strong>multtest</strong>
 2813 
<br><dt><strong>multtest</strong>
2809 
     <dd>Multiple testing procedures for controlling the family-wise error rate
 2814 
     <dd>Multiple testing procedures for controlling the family-wise error rate
2810 
(FWER) and the false discovery rate (FDR).  Tests can be based on
 2815 
(FWER) and the false discovery rate (FDR).  Tests can be based on
2811 
t- or F-statistics for one- and two-factor designs, and
 2816 
t- or F-statistics for one- and two-factor designs, and
2812 
permutation procedures are available to estimate adjusted
 2817 
permutation procedures are available to estimate adjusted
2813 
p-values.
2818 
p-values.
2814 
<br><dt><strong>ontoTools</strong>
 2819 
<br><dt><strong>ontoTools</strong>
2815 
     <dd>Graphs and sparse matrices for working with ontologies.
2820 
     <dd>Graphs and sparse matrices for working with ontologies.
2816 
<br><dt><strong>pamr</strong>
 2821 
<br><dt><strong>pamr</strong>
2817 
     <dd>Pam: prediction analysis for microarrays.
2822 
     <dd>Pam: prediction analysis for microarrays.
2818 
<br><dt><strong>reposTools</strong>
 2823 
<br><dt><strong>reposTools</strong>
2819 
     <dd>Tools for dealing with file repositories and allow users to easily
 2824 
     <dd>Tools for dealing with file repositories and allow users to easily
2820 
install, update, and distribute packages, vignettes, and other files.
2825 
install, update, and distribute packages, vignettes, and other files.
2821 
<br><dt><strong>rhdf5</strong>
 2826 
<br><dt><strong>rhdf5</strong>
2822 
     <dd>Storage and retrieval of large datasets using the HDF5 library and file
 2827 
     <dd>Storage and retrieval of large datasets using the HDF5 library and file
2823 
format.
2828 
format.
2824 
<br><dt><strong>siggenes</strong>
 2829 
<br><dt><strong>siggenes</strong>
2825 
     <dd>Identifying differentially expressed genes and estimating the False
 2830 
     <dd>Identifying differentially expressed genes and estimating the False
2826 
Discovery Rate (FDR) with both the Significance Analysis of Microarrays
 2831 
Discovery Rate (FDR) with both the Significance Analysis of Microarrays
2827 
(SAM) and the Empirical Bayes Analyses of Microarrays (EBAM).
2832 
(SAM) and the Empirical Bayes Analyses of Microarrays (EBAM).
2828 
<br><dt><strong>splicegear</strong>
 2833 
<br><dt><strong>splicegear</strong>
2829 
     <dd>A set of tools to work with alternative splicing.
2834 
     <dd>A set of tools to work with alternative splicing.
2830 
<br><dt><strong>tkWidgets</strong>
 2835 
<br><dt><strong>tkWidgets</strong>
2831 
     <dd>Widgets in Tcl/Tk that provide functionality for Bioconductor packages.
2836 
     <dd>Widgets in Tcl/Tk that provide functionality for Bioconductor packages.
2832 
<br><dt><strong>vsn</strong>
 2837 
<br><dt><strong>vsn</strong>
2833 
     <dd>Calibration and variance stabilizing transformations for both Affymetrix
 2838 
     <dd>Calibration and variance stabilizing transformations for both Affymetrix
2834 
and cDNA array data.
2839 
and cDNA array data.
2835 
<br><dt><strong>widgetTools</strong>
 2840 
<br><dt><strong>widgetTools</strong>
2836 
     <dd>Tools for creating Tcl/Tk widgets, i.e., small-scale graphical user
 2841 
     <dd>Tools for creating Tcl/Tk widgets, i.e., small-scale graphical user
2837 
interfaces.
2842 
interfaces.
2838 
</dl>
 2843 
</dl>
2839 
 
 2844 
 
2840 
<div class="node">
 2845 
<div class="node">
2841 
<p><hr>
 2846 
<p><hr>
2842 
Node:&nbsp;<a name="Other%20add-on%20packages">Other add-on packages</a>,
 2847 
Node:&nbsp;<a name="Other%20add-on%20packages">Other add-on packages</a>,
2843 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
 2848 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Add-on%20packages%20from%20BioConductor">Add-on packages from BioConductor</a>,
2844 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
 2849 
Up:&nbsp;<a rel="up" accesskey="u" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>
2845 
<br>
 2850 
<br>
2846 
</div>
 2851 
</div>
2847 
 
 2852 
 
2848 
<h3 class="subsection">5.1.5 Other add-on packages</h4>
 2853 
<h3 class="subsection">5.1.5 Other add-on packages</h4>
2849 
 
 2854 
 
2850 
<a href="mailto:jlindsey@luc.ac.be">Jim Lindsey</a> has written a collection of R
 2855 
<a href="mailto:jlindsey@luc.ac.be">Jim Lindsey</a> has written a collection of R
2851 
packages for nonlinear regression and repeated measurements, consisting
 2856 
packages for nonlinear regression and repeated measurements, consisting
2852 
of <strong>event</strong> (event history procedures and models), <strong>gnlm</strong>
 2857 
of <strong>event</strong> (event history procedures and models), <strong>gnlm</strong>
2853 
(generalized nonlinear regression models), <strong>growth</strong> (multivariate
 2858 
(generalized nonlinear regression models), <strong>growth</strong> (multivariate
2854 
normal and elliptically-contoured repeated measurements models),
 2859 
normal and elliptically-contoured repeated measurements models),
2855 
<strong>repeated</strong> (non-normal repeated measurements models),
 2860 
<strong>repeated</strong> (non-normal repeated measurements models),
2856 
<strong>rmutil</strong> (utilities for nonlinear regression and repeated
 2861 
<strong>rmutil</strong> (utilities for nonlinear regression and repeated
2857 
measurements), and <strong>stable</strong> (probability functions and
 2862 
measurements), and <strong>stable</strong> (probability functions and
2858 
generalized regression models for stable distributions).  All analyses
 2863 
generalized regression models for stable distributions).  All analyses
2859 
in the new edition of his book "Models for Repeated Measurements"
 2864 
in the new edition of his book "Models for Repeated Measurements"
2860 
(1999, Oxford University Press) were carried out using these packages.
2865 
(1999, Oxford University Press) were carried out using these packages.
2861 
Jim has also started <strong>dna</strong>, a package with procedures for the
 2866 
Jim has also started <strong>dna</strong>, a package with procedures for the
2862 
analysis of DNA sequences.  Jim's packages can be obtained from
 2867 
analysis of DNA sequences.  Jim's packages can be obtained from
2863 
<a href="http://www.luc.ac.be/~jlindsey/rcode.html">http://www.luc.ac.be/~jlindsey/rcode.html</a>.
 2868 
<a href="http://www.luc.ac.be/~jlindsey/rcode.html">http://www.luc.ac.be/~jlindsey/rcode.html</a>.
2864 
 
 2869 
 
2865 
   <p>More code has been posted to the R-help mailing list, and can be
 2870 
   <p>More code has been posted to the R-help mailing list, and can be
2866 
obtained from the mailing list archive.
 2871 
obtained from the mailing list archive.
2867 
 
 2872 
 
2868 
<div class="node">
 2873 
<div class="node">
2869 
<p><hr>
 2874 
<p><hr>
2870 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
 2875 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
2871 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
 2876 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
2872 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
 2877 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
2873 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 2878 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
2874 
<br>
 2879 
<br>
2875 
</div>
 2880 
</div>
2876 
 
 2881 
 
2877 
<h3 class="section">5.2 How can add-on packages be installed?</h3>
 2882 
<h3 class="section">5.2 How can add-on packages be installed?</h3>
2878 
 
 2883 
 
2879 
<p>(Unix only.)  The add-on packages on <small>CRAN</small> come as gzipped tar
 2884 
<p>(Unix only.)  The add-on packages on <small>CRAN</small> come as gzipped tar
2880 
files named <code></code><var>pkg</var><code>_</code><var>version</var><code>.tar.gz</code>, which may in fact be
 2885 
files named <code></code><var>pkg</var><code>_</code><var>version</var><code>.tar.gz</code>, which may in fact be
2881 
"bundles" containing more than one package.  Provided that
 2886 
"bundles" containing more than one package.  Provided that
2882 
<code>tar</code> and <code>gzip</code> are available on your system, type
 2887 
<code>tar</code> and <code>gzip</code> are available on your system, type
2883 
 
 2888 
 
2884 
<pre class="example">     $ R CMD INSTALL /path/to/<var>pkg</var>_<var>version</var>.tar.gz
 2889 
<pre class="example">     $ R CMD INSTALL /path/to/<var>pkg</var>_<var>version</var>.tar.gz
2885 
     </pre>
 2890 
     </pre>
2886 
 
 2891 
 
2887 
<p>at the shell prompt to install to the library tree rooted at the first
 2892 
<p>at the shell prompt to install to the library tree rooted at the first
2888 
directory given in <code>R_LIBS</code> (see below) if this is set and non-null,
 2893 
directory given in <code>R_LIBS</code> (see below) if this is set and non-null,
2889 
and to the default library (the <code>library</code> subdirectory of
 2894 
and to the default library (the <code>library</code> subdirectory of
2890 
<code>R_HOME</code>) otherwise.  (Versions of R prior to 1.3.0 installed
 2895 
<code>R_HOME</code>) otherwise.  (Versions of R prior to 1.3.0 installed
2891 
to the default library by default.)
 2896 
to the default library by default.)
2892 
 
 2897 
 
2893 
   <p>To install to another tree (e.g., your private one), use
 2898 
   <p>To install to another tree (e.g., your private one), use
2894 
 
 2899 
 
2895 
<pre class="example">     $ R CMD INSTALL -l <var>lib</var> /path/to/<var>pkg</var>_<var>version</var>.tar.gz
 2900 
<pre class="example">     $ R CMD INSTALL -l <var>lib</var> /path/to/<var>pkg</var>_<var>version</var>.tar.gz
2896 
     </pre>
 2901 
     </pre>
2897 
 
 2902 
 
2898 
<p>where <var>lib</var> gives the path to the library tree to install to.
 2903 
<p>where <var>lib</var> gives the path to the library tree to install to.
2899 
 
 2904 
 
2900 
   <p>Even more conveniently, you can install and automatically update
 2905 
   <p>Even more conveniently, you can install and automatically update
2901 
packages from within R if you have access to <small>CRAN</small>.  See the
 2906 
packages from within R if you have access to <small>CRAN</small>.  See the
2902 
help page for <code>CRAN.packages()</code> for more information.
 2907 
help page for <code>CRAN.packages()</code> for more information.
2903 
 
 2908 
 
2904 
   <p>You can use several library trees of add-on packages.  The easiest way
 2909 
   <p>You can use several library trees of add-on packages.  The easiest way
2905 
to tell R to use these is via the environment variable <code>R_LIBS</code>
 2910 
to tell R to use these is via the environment variable <code>R_LIBS</code>
2906 
which should be a colon-separated list of directories at which R library
 2911 
which should be a colon-separated list of directories at which R library
2907 
trees are rooted.  You do not have to specify the default tree in
 2912 
trees are rooted.  You do not have to specify the default tree in
2908 
<code>R_LIBS</code>.  E.g., to use a private tree in <code>$HOME/lib/R</code> and a
 2913 
<code>R_LIBS</code>.  E.g., to use a private tree in <code>$HOME/lib/R</code> and a
2909 
public site-wide tree in <code>/usr/local/lib/R-contrib</code>, put
 2914 
public site-wide tree in <code>/usr/local/lib/R-contrib</code>, put
2910 
 
 2915 
 
2911 
<pre class="example">     R_LIBS="$HOME/lib/R:/usr/local/lib/R-contrib"; export R_LIBS
 2916 
<pre class="example">     R_LIBS="$HOME/lib/R:/usr/local/lib/R-contrib"; export R_LIBS
2912 
     </pre>
 2917 
     </pre>
2913 
 
 2918 
 
2914 
<p>into your (Bourne) shell profile or even preferably, add the line
 2919 
<p>into your (Bourne) shell profile or even preferably, add the line
2915 
 
 2920 
 
2916 
<pre class="example">     R_LIBS="$HOME/lib/R:/usr/local/lib/R-contrib"
 2921 
<pre class="example">     R_LIBS="$HOME/lib/R:/usr/local/lib/R-contrib"
2917 
     </pre>
 2922 
     </pre>
2918 
 
 2923 
 
2919 
<p>your <code>~/.Renviron</code> file.  (Note that no <code>export</code> statement is
 2924 
<p>your <code>~/.Renviron</code> file.  (Note that no <code>export</code> statement is
2920 
needed or allowed in this file; see the on-line help for <code>Startup</code>
 2925 
needed or allowed in this file; see the on-line help for <code>Startup</code>
2921 
for more information.)
 2926 
for more information.)
2922 
 
 2927 
 
2923 
<div class="node">
 2928 
<div class="node">
2924 
<p><hr>
 2929 
<p><hr>
2925 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
 2930 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
2926 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
 2931 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
2927 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
 2932 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
2928 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 2933 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
2929 
<br>
 2934 
<br>
2930 
</div>
 2935 
</div>
2931 
 
 2936 
 
2932 
<h3 class="section">5.3 How can add-on packages be used?</h3>
 2937 
<h3 class="section">5.3 How can add-on packages be used?</h3>
2933 
 
 2938 
 
2934 
<p>To find out which additional packages are available on your system, type
 2939 
<p>To find out which additional packages are available on your system, type
2935 
 
 2940 
 
2936 
<pre class="example">     library()
 2941 
<pre class="example">     library()
2937 
     </pre>
 2942 
     </pre>
2938 
 
 2943 
 
2939 
<p>at the R prompt.
 2944 
<p>at the R prompt.
2940 
 
 2945 
 
2941 
   <p>This produces something like
 2946 
   <p>This produces something like
2942 
 
 2947 
 
2943 
<pre class="smallexample">     Packages in `/home/me/lib/R':
 2948 
<pre class="smallexample">     Packages in `/home/me/lib/R':
2944
2949
2945 
     mystuff       My own R functions, nicely packaged but not documented
 2950 
     mystuff       My own R functions, nicely packaged but not documented
2946
2951
2947 
     Packages in `/usr/local/lib/R/library':
 2952 
     Packages in `/usr/local/lib/R/library':
2948
2953
2949 
     KernSmooth    Functions for kernel smoothing for Wand &amp; Jones (1995)
 2954 
     KernSmooth    Functions for kernel smoothing for Wand &amp; Jones (1995)
2950 
     MASS          Main Library of Venables and Ripley's MASS
 2955 
     MASS          Main Library of Venables and Ripley's MASS
2951 
     base          The R base package
 2956 
     base          The R base package
2952 
     boot          Bootstrap R (S-Plus) Functions (Canty)
 2957 
     boot          Bootstrap R (S-Plus) Functions (Canty)
2953 
     class         Functions for classification
 2958 
     class         Functions for classification
2954 
     cluster       Functions for clustering (by Rousseeuw et al.)
 2959 
     cluster       Functions for clustering (by Rousseeuw et al.)
2955 
     ctest         Classical Tests
 2960 
     ctest         Classical Tests
2956 
     eda           Exploratory Data Analysis
 2961 
     eda           Exploratory Data Analysis
2957 
     foreign       Read data stored by Minitab, S, SAS, SPSS, Stata, ...
 2962 
     foreign       Read data stored by Minitab, S, SAS, SPSS, Stata, ...
2958 
     grid          The Grid Graphics Package
 2963 
     grid          The Grid Graphics Package
2959 
     lattice       Lattice Graphics
 2964 
     lattice       Lattice Graphics
2960 
     lqs           Resistant Regression and Covariance Estimation
 2965 
     lqs           Resistant Regression and Covariance Estimation
2961 
     methods       Formal Methods and Classes
 2966 
     methods       Formal Methods and Classes
2962 
     mle           Maximum likelihood estimation
 2967 
     mle           Maximum likelihood estimation
2963 
     mgcv          Multiple smoothing parameter estimation and GAMs by GCV
 2968 
     mgcv          Multiple smoothing parameter estimation and GAMs by GCV
2964 
     modreg        Modern Regression: Smoothing and Local Methods
 2969 
     modreg        Modern Regression: Smoothing and Local Methods
2965 
     mva           Classical Multivariate Analysis
 2970 
     mva           Classical Multivariate Analysis
2966 
     nlme          Linear and nonlinear mixed effects models
 2971 
     nlme          Linear and nonlinear mixed effects models
2967 
     nls           Nonlinear regression
 2972 
     nls           Nonlinear regression
2968 
     nnet          Feed-forward neural networks and multinomial log-linear
 2973 
     nnet          Feed-forward neural networks and multinomial log-linear
2969 
                   models
 2974 
                   models
2970 
     rpart         Recursive partitioning
 2975 
     rpart         Recursive partitioning
2971 
     spatial       functions for kriging and point pattern analysis
 2976 
     spatial       functions for kriging and point pattern analysis
2972 
     splines       Regression Spline Functions and Classes
 2977 
     splines       Regression Spline Functions and Classes
2973 
     stepfun       Step Functions, including Empirical Distributions
 2978 
     stepfun       Step Functions, including Empirical Distributions
2974 
     survival      Survival analysis, including penalised likelihood
 2979 
     survival      Survival analysis, including penalised likelihood
2975 
     tcltk         Tcl/Tk Interface
 2980 
     tcltk         Tcl/Tk Interface
2976 
     tools         Tools for Package Development and Administration
 2981 
     tools         Tools for Package Development and Administration
2977 
     ts            Time series functions
 2982 
     ts            Time series functions
2978 
     </pre>
 2983 
     </pre>
2979 
 
 2984 
 
2980 
   <p>You can "load" the installed package <var>pkg</var> by
 2985 
   <p>You can "load" the installed package <var>pkg</var> by
2981 
 
 2986 
 
2982 
<pre class="example">     library(<var>pkg</var>)
 2987 
<pre class="example">     library(<var>pkg</var>)
2983 
     </pre>
 2988 
     </pre>
2984 
 
 2989 
 
2985 
   <p>You can then find out which functions it provides by typing one of
 2990 
   <p>You can then find out which functions it provides by typing one of
2986 
 
 2991 
 
2987 
<pre class="example">     library(help = <var>pkg</var>)
 2992 
<pre class="example">     library(help = <var>pkg</var>)
2988 
     help(package = <var>pkg</var>)
 2993 
     help(package = <var>pkg</var>)
2989 
     </pre>
 2994 
     </pre>
2990 
 
 2995 
 
2991 
   <p>You can unload the loaded package <var>pkg</var> by
 2996 
   <p>You can unload the loaded package <var>pkg</var> by
2992 
 
 2997 
 
2993 
<pre class="example">     detach("package:<var>pkg</var>")
 2998 
<pre class="example">     detach("package:<var>pkg</var>")
2994 
     </pre>
 2999 
     </pre>
2995 
 
 3000 
 
2996 
<div class="node">
 3001 
<div class="node">
2997 
<p><hr>
 3002 
<p><hr>
2998 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
 3003 
Node:&nbsp;<a name="How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
2999 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
 3004 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
3000 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
 3005 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
3001 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 3006 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
3002 
<br>
 3007 
<br>
3003 
</div>
 3008 
</div>
3004 
 
 3009 
 
3005 
<h3 class="section">5.4 How can add-on packages be removed?</h3>
 3010 
<h3 class="section">5.4 How can add-on packages be removed?</h3>
3006 
 
 3011 
 
3007 
<p>Use
 3012 
<p>Use
3008 
 
 3013 
 
3009 
<pre class="example">     $ R CMD REMOVE <var>pkg_1</var> ... <var>pkg_n</var>
 3014 
<pre class="example">     $ R CMD REMOVE <var>pkg_1</var> ... <var>pkg_n</var>
3010 
     </pre>
 3015 
     </pre>
3011 
 
 3016 
 
3012 
<p>to remove the packages <var>pkg_1</var>, <small class="dots">...</small>, <var>pkg_n</var> from the
 3017 
<p>to remove the packages <var>pkg_1</var>, <small class="dots">...</small>, <var>pkg_n</var> from the
3013 
library tree rooted at the first directory given in <code>R_LIBS</code> if this
 3018 
library tree rooted at the first directory given in <code>R_LIBS</code> if this
3014 
is set and non-null, and from the default library otherwise.  (Versions
 3019 
is set and non-null, and from the default library otherwise.  (Versions
3015 
of R prior to 1.3.0 removed from the default library by default.)
 3020 
of R prior to 1.3.0 removed from the default library by default.)
3016 
 
 3021 
 
3017 
   <p>To remove from library <var>lib</var>, do
 3022 
   <p>To remove from library <var>lib</var>, do
3018 
 
 3023 
 
3019 
<pre class="example">     $ R CMD REMOVE -l <var>lib</var> <var>pkg_1</var> ... <var>pkg_n</var>
 3024 
<pre class="example">     $ R CMD REMOVE -l <var>lib</var> <var>pkg_1</var> ... <var>pkg_n</var>
3020 
     </pre>
 3025 
     </pre>
3021 
 
 3026 
 
3022 
<div class="node">
 3027 
<div class="node">
3023 
<p><hr>
 3028 
<p><hr>
3024 
Node:&nbsp;<a name="How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
 3029 
Node:&nbsp;<a name="How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
3025 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
 3030 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
3026 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
 3031 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
3027 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 3032 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
3028 
<br>
 3033 
<br>
3029 
</div>
 3034 
</div>
3030 
 
 3035 
 
3031 
<h3 class="section">5.5 How can I create an R package?</h3>
 3036 
<h3 class="section">5.5 How can I create an R package?</h3>
3032 
 
 3037 
 
3033 
<p>A package consists of a subdirectory containing the files
 3038 
<p>A package consists of a subdirectory containing the files
3034 
<code>DESCRIPTION</code> and <code>INDEX</code>, and the subdirectories <code>R</code>,
 3039 
<code>DESCRIPTION</code> and <code>INDEX</code>, and the subdirectories <code>R</code>,
3035 
<code>data</code>, <code>demo</code>, <code>exec</code>, <code>inst</code>, <code>man</code>,
 3040 
<code>data</code>, <code>demo</code>, <code>exec</code>, <code>inst</code>, <code>man</code>,
3036 
<code>src</code>, and <code>tests</code> (some of which can be missing).  Optionally
 3041 
<code>src</code>, and <code>tests</code> (some of which can be missing).  Optionally
3037 
the package can also contain script files <code>configure</code> and
 3042 
the package can also contain script files <code>configure</code> and
3038 
<code>cleanup</code> which are executed before and after installation.
 3043 
<code>cleanup</code> which are executed before and after installation.
3039 
 
 3044 
 
3040 
   <p>See section "Creating R packages" in <cite>Writing R Extensions</cite>, for
 3045 
   <p>See section "Creating R packages" in <cite>Writing R Extensions</cite>, for
3041 
details.
3046 
details.
3042 
This manual is included in the R distribution, see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>, and gives information on package structure, the
 3047 
This manual is included in the R distribution, see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>, and gives information on package structure, the
3043 
configure and cleanup mechanisms, and on automated package checking and
 3048 
configure and cleanup mechanisms, and on automated package checking and
3044 
building.
 3049 
building.
3045 
 
 3050 
 
3046 
   <p>R version 1.3.0 has added the function <code>package.skeleton()</code> which
 3051 
   <p>R version 1.3.0 has added the function <code>package.skeleton()</code> which
3047 
will set up directories, save data and code, and create skeleton help
 3052 
will set up directories, save data and code, and create skeleton help
3048 
files for a set of R functions and datasets.
 3053 
files for a set of R functions and datasets.
3049 
 
 3054 
 
3050 
   <p>See <a href="#What%20is%20CRAN%3f">What is CRAN?</a>, for information on uploading a package to <small>CRAN</small>.
 3055 
   <p>See <a href="#What%20is%20CRAN%3f">What is CRAN?</a>, for information on uploading a package to <small>CRAN</small>.
3051 
 
 3056 
 
3052 
<div class="node">
 3057 
<div class="node">
3053 
<p><hr>
 3058 
<p><hr>
3054 
Node:&nbsp;<a name="How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
 3059 
Node:&nbsp;<a name="How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
3055 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
 3060 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
3056 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
 3061 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Add-On%20Packages">R Add-On Packages</a>
3057 
<br>
 3062 
<br>
3058 
</div>
 3063 
</div>
3059 
 
 3064 
 
3060 
<h3 class="section">5.6 How can I contribute to R?</h3>
 3065 
<h3 class="section">5.6 How can I contribute to R?</h3>
3061 
 
 3066 
 
3062 
<p>R is in active development and there is always a risk of bugs creeping
 3067 
<p>R is in active development and there is always a risk of bugs creeping
3063 
in.  Also, the developers do not have access to all possible machines
 3068 
in.  Also, the developers do not have access to all possible machines
3064 
capable of running R.  So, simply using it and communicating problems is
 3069 
capable of running R.  So, simply using it and communicating problems is
3065 
certainly of great value.
 3070 
certainly of great value.
3066 
 
 3071 
 
3067 
   <p>One place where functionality is still missing is the modeling software
 3072 
   <p>One place where functionality is still missing is the modeling software
3068 
as described in "Statistical Models in S" (see <a href="#What%20is%20S%3f">What is S?</a>);
 3073 
as described in "Statistical Models in S" (see <a href="#What%20is%20S%3f">What is S?</a>);
3069 
Generalized Additive Models (see <a href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>) and
 3074 
Generalized Additive Models (see <a href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>) and
3070 
some of the nonlinear modeling code are not there yet.
 3075 
some of the nonlinear modeling code are not there yet.
3071 
 
 3076 
 
3072 
   <p>The <a href="http://developer.R-project.org/">R Developer Page</a> acts as an
 3077 
   <p>The <a href="http://developer.R-project.org/">R Developer Page</a> acts as an
3073 
intermediate repository for more or less finalized ideas and plans for
 3078 
intermediate repository for more or less finalized ideas and plans for
3074 
the R statistical system.  It contains (pointers to) TODO lists, RFCs,
 3079 
the R statistical system.  It contains (pointers to) TODO lists, RFCs,
3075 
various other writeups, ideas lists, and CVS miscellanea.
 3080 
various other writeups, ideas lists, and CVS miscellanea.
3076 
 
 3081 
 
3077 
   <p>Many (more) of the packages available at the Statlib S Repository might
 3082 
   <p>Many (more) of the packages available at the Statlib S Repository might
3078 
be worth porting to R.
 3083 
be worth porting to R.
3079 
 
 3084 
 
3080 
   <p>If you are interested in working on any of these projects, please notify
 3085 
   <p>If you are interested in working on any of these projects, please notify
3081 
<a href="mailto:Kurt.Hornik@R-project.org">Kurt Hornik</a>.
 3086 
<a href="mailto:Kurt.Hornik@R-project.org">Kurt Hornik</a>.
3082 
 
 3087 
 
3083 
<div class="node">
 3088 
<div class="node">
3084 
<p><hr>
 3089 
<p><hr>
3085 
Node:&nbsp;<a name="R%20and%20Emacs">R and Emacs</a>,
 3090 
Node:&nbsp;<a name="R%20and%20Emacs">R and Emacs</a>,
3086 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Miscellanea">R Miscellanea</a>,
 3091 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Miscellanea">R Miscellanea</a>,
3087 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
 3092 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Add-On%20Packages">R Add-On Packages</a>,
3088 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 3093 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
3089 
<br>
 3094 
<br>
3090 
</div>
 3095 
</div>
3091 
 
 3096 
 
3092 
<h2 class="chapter">6 R and Emacs</h2>
 3097 
<h2 class="chapter">6 R and Emacs</h2>
3093 
 
 3098 
 
3094 
<ul class="menu">
 3099 
<ul class="menu">
3095 
<li><a accesskey="1" href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>:
3100 
<li><a accesskey="1" href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>:
3096 
<li><a accesskey="2" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>:
3101 
<li><a accesskey="2" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>:
3097 
<li><a accesskey="3" href="#Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>:
3102 
<li><a accesskey="3" href="#Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>:
3098 
</ul>
 3103 
</ul>
3099 
 
 3104 
 
3100 
<div class="node">
 3105 
<div class="node">
3101 
<p><hr>
 3106 
<p><hr>
3102 
Node:&nbsp;<a name="Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
 3107 
Node:&nbsp;<a name="Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
3103 
Next:&nbsp;<a rel="next" accesskey="n" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
 3108 
Next:&nbsp;<a rel="next" accesskey="n" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
3104 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20Emacs">R and Emacs</a>,
 3109 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20Emacs">R and Emacs</a>,
3105 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
 3110 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
3106 
<br>
 3111 
<br>
3107 
</div>
 3112 
</div>
3108 
 
 3113 
 
3109 
<h3 class="section">6.1 Is there Emacs support for R?</h3>
 3114 
<h3 class="section">6.1 Is there Emacs support for R?</h3>
3110 
 
 3115 
 
3111 
<p>There is an Emacs package called <small>ESS</small> ("Emacs Speaks
 3116 
<p>There is an Emacs package called <small>ESS</small> ("Emacs Speaks
3112 
Statistics") which provides a standard interface between statistical
 3117 
Statistics") which provides a standard interface between statistical
3113 
programs and statistical processes.  It is intended to provide
 3118 
programs and statistical processes.  It is intended to provide
3114 
assistance for interactive statistical programming and data analysis.
3119 
assistance for interactive statistical programming and data analysis.
3115 
Languages supported include: S dialects (S 3/4, <small>S-PLUS</small> 3.x/4.x/5.x,
 3120 
Languages supported include: S dialects (S 3/4, <small>S-PLUS</small> 3.x/4.x/5.x,
3116 
and R), LispStat dialects (XLispStat, ViSta) and SAS.  Stata and SPSS
 3121 
and R), LispStat dialects (XLispStat, ViSta) and SAS.  Stata and SPSS
3117 
dialect (SPSS, PSPP) support is being examined for possible future
 3122 
dialect (SPSS, PSPP) support is being examined for possible future
3118 
implementation
 3123 
implementation
3119 
 
 3124 
 
3120 
   <p><small>ESS</small> grew out of the need for bug fixes and extensions to
 3125 
   <p><small>ESS</small> grew out of the need for bug fixes and extensions to
3121 
S-mode 4.8 (which was a <small>GNU</small> Emacs interface to S/<small>S-PLUS</small>
 3126 
S-mode 4.8 (which was a <small>GNU</small> Emacs interface to S/<small>S-PLUS</small>
3122 
version 3 only).  The current set of developers desired support for
 3127 
version 3 only).  The current set of developers desired support for
3123 
XEmacs, R, S4, and MS Windows.  In addition, with new modes being
 3128 
XEmacs, R, S4, and MS Windows.  In addition, with new modes being
3124 
developed for R, Stata, and SAS, it was felt that a unifying interface
 3129 
developed for R, Stata, and SAS, it was felt that a unifying interface
3125 
and framework for the user interface would benefit both the user and the
 3130 
and framework for the user interface would benefit both the user and the
3126 
developer, by helping both groups conform to standard Emacs usage.  The
 3131 
developer, by helping both groups conform to standard Emacs usage.  The
3127 
end result is an increase in efficiency for statistical programming and
 3132 
end result is an increase in efficiency for statistical programming and
3128 
data analysis, over the usual tools.
 3133 
data analysis, over the usual tools.
3129 
 
 3134 
 
3130 
   <p>R support contains code for editing R source code (syntactic indentation
 3135 
   <p>R support contains code for editing R source code (syntactic indentation
3131 
and highlighting of source code, partial evaluations of code, loading
 3136 
and highlighting of source code, partial evaluations of code, loading
3132 
and error-checking of code, and source code revision maintenance) and
 3137 
and error-checking of code, and source code revision maintenance) and
3133 
documentation (syntactic indentation and highlighting of source code,
 3138 
documentation (syntactic indentation and highlighting of source code,
3134 
sending examples to running <small>ESS</small> process, and previewing),
 3139 
sending examples to running <small>ESS</small> process, and previewing),
3135 
interacting with an inferior R process from within Emacs (command-line
 3140 
interacting with an inferior R process from within Emacs (command-line
3136 
editing, searchable command history, command-line completion of R object
 3141 
editing, searchable command history, command-line completion of R object
3137 
and file names, quick access to object and search lists, transcript
 3142 
and file names, quick access to object and search lists, transcript
3138 
recording, and an interface to the help system), and transcript
 3143 
recording, and an interface to the help system), and transcript
3139 
manipulation (recording and saving transcript files, manipulating and
 3144 
manipulation (recording and saving transcript files, manipulating and
3140 
editing saved transcripts, and re-evaluating commands from transcript
 3145 
editing saved transcripts, and re-evaluating commands from transcript
3141 
files).
 3146 
files).
3142 
 
 3147 
 
3143 
   <p>The latest stable version of <small>ESS</small> are available via <small>CRAN</small> or
 3148 
   <p>The latest stable version of <small>ESS</small> are available via <small>CRAN</small> or
3144 
the <a href="http://ESS.R-project.org/">ESS web page</a>.  The <small>HTML</small> version
 3149 
the <a href="http://ESS.R-project.org/">ESS web page</a>.  The <small>HTML</small> version
3145 
of the documentation can be found at <a href="http://stat.ethz.ch/ESS/">http://stat.ethz.ch/ESS/</a>.
 3150 
of the documentation can be found at <a href="http://stat.ethz.ch/ESS/">http://stat.ethz.ch/ESS/</a>.
3146 
 
 3151 
 
3147 
   <p><small>ESS</small> comes with detailed installation instructions.
 3152 
   <p><small>ESS</small> comes with detailed installation instructions.
3148 
 
 3153 
 
3149 
   <p>For help with <small>ESS</small>, send email to
 3154 
   <p>For help with <small>ESS</small>, send email to
3150 
<a href="mailto:ESS-help@stat.ethz.ch">ESS-help@stat.ethz.ch</a>.
 3155 
<a href="mailto:ESS-help@stat.ethz.ch">ESS-help@stat.ethz.ch</a>.
3151 
 
 3156 
 
3152 
   <p>Please send bug reports and suggestions on <small>ESS</small> to
 3157 
   <p>Please send bug reports and suggestions on <small>ESS</small> to
3153 
<a href="mailto:ESS-bugs@stat.math.ethz.ch">ESS-bugs@stat.math.ethz.ch</a>.  The easiest way to do this from is
 3158 
<a href="mailto:ESS-bugs@stat.math.ethz.ch">ESS-bugs@stat.math.ethz.ch</a>.  The easiest way to do this from is
3154 
within Emacs by typing <kbd>M-x ess-submit-bug-report</kbd> or using the
 3159 
within Emacs by typing <kbd>M-x ess-submit-bug-report</kbd> or using the
3155 
[ESS] or [iESS] pulldown menus.
 3160 
[ESS] or [iESS] pulldown menus.
3156 
 
 3161 
 
3157 
<div class="node">
 3162 
<div class="node">
3158 
<p><hr>
 3163 
<p><hr>
3159 
Node:&nbsp;<a name="Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
 3164 
Node:&nbsp;<a name="Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
3160 
Next:&nbsp;<a rel="next" accesskey="n" href="#Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>,
 3165 
Next:&nbsp;<a rel="next" accesskey="n" href="#Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>,
3161 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
 3166 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
3162 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
 3167 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
3163 
<br>
 3168 
<br>
3164 
</div>
 3169 
</div>
3165 
 
 3170 
 
3166 
<h3 class="section">6.2 Should I run R from within Emacs?</h3>
 3171 
<h3 class="section">6.2 Should I run R from within Emacs?</h3>
3167 
 
 3172 
 
3168 
<p>Yes, <em>definitely</em>.  Inferior R mode provides a readline/history
 3173 
<p>Yes, <em>definitely</em>.  Inferior R mode provides a readline/history
3169 
mechanism, object name completion, and syntax-based highlighting of the
 3174 
mechanism, object name completion, and syntax-based highlighting of the
3170 
interaction buffer using Font Lock mode, as well as a very convenient
 3175 
interaction buffer using Font Lock mode, as well as a very convenient
3171 
interface to the R help system.
 3176 
interface to the R help system.
3172 
 
 3177 
 
3173 
   <p>Of course, it also integrates nicely with the mechanisms for editing R
 3178 
   <p>Of course, it also integrates nicely with the mechanisms for editing R
3174 
source using Emacs.  One can write code in one Emacs buffer and send
 3179 
source using Emacs.  One can write code in one Emacs buffer and send
3175 
whole or parts of it for execution to R; this is helpful for both data
 3180 
whole or parts of it for execution to R; this is helpful for both data
3176 
analysis and programming.  One can also seamlessly integrate with a
 3181 
analysis and programming.  One can also seamlessly integrate with a
3177 
revision control system, in order to maintain a log of changes in your
 3182 
revision control system, in order to maintain a log of changes in your
3178 
programs and data, as well as to allow for the retrieval of past
 3183 
programs and data, as well as to allow for the retrieval of past
3179 
versions of the code.
 3184 
versions of the code.
3180 
 
 3185 
 
3181 
   <p>In addition, it allows you to keep a record of your session, which can
 3186 
   <p>In addition, it allows you to keep a record of your session, which can
3182 
also be used for error recovery through the use of the transcript mode.
 3187 
also be used for error recovery through the use of the transcript mode.
3183 
 
 3188 
 
3184 
   <p>To specify command line arguments for the inferior R process, use
 3189 
   <p>To specify command line arguments for the inferior R process, use
3185 
<kbd>C-u M-x R</kbd> for starting R.
 3190 
<kbd>C-u M-x R</kbd> for starting R.
3186 
 
 3191 
 
3187 
<div class="node">
 3192 
<div class="node">
3188 
<p><hr>
 3193 
<p><hr>
3189 
Node:&nbsp;<a name="Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>,
 3194 
Node:&nbsp;<a name="Debugging%20R%20from%20within%20Emacs">Debugging R from within Emacs</a>,
3190 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
 3195 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
3191 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
 3196 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20and%20Emacs">R and Emacs</a>
3192 
<br>
 3197 
<br>
3193 
</div>
 3198 
</div>
3194 
 
 3199 
 
3195 
<h3 class="section">6.3 Debugging R from within Emacs</h3>
 3200 
<h3 class="section">6.3 Debugging R from within Emacs</h3>
3196 
 
 3201 
 
3197 
<p>To debug R "from within Emacs", there are several possibilities.  To
 3202 
<p>To debug R "from within Emacs", there are several possibilities.  To
3198 
use the Emacs GUD (Grand Unified Debugger) library with the recommended
 3203 
use the Emacs GUD (Grand Unified Debugger) library with the recommended
3199 
debugger GDB, type <kbd>M-x gdb</kbd> and give the path to the R
 3204 
debugger GDB, type <kbd>M-x gdb</kbd> and give the path to the R
3200 
<em>binary</em> as argument.  At the <code>gdb</code> prompt, set
 3205 
<em>binary</em> as argument.  At the <code>gdb</code> prompt, set
3201 
<code>R_HOME</code> and other environment variables as needed (using e.g.
3206 
<code>R_HOME</code> and other environment variables as needed (using e.g.
3202 
<kbd>set env R_HOME /path/to/R/</kbd>, but see also below), and start the
 3207 
<kbd>set env R_HOME /path/to/R/</kbd>, but see also below), and start the
3203 
binary with the desired arguments (e.g., <kbd>run --quiet</kbd>).
 3208 
binary with the desired arguments (e.g., <kbd>run --quiet</kbd>).
3204 
 
 3209 
 
3205 
   <p>If you have <small>ESS</small>, you can do <kbd>C-u M-x R &lt;RET&gt; - d
 3210 
   <p>If you have <small>ESS</small>, you can do <kbd>C-u M-x R &lt;RET&gt; - d
3206 
&lt;SPC&gt; g d b &lt;RET&gt;</kbd> to start an inferior R process with arguments
 3211 
&lt;SPC&gt; g d b &lt;RET&gt;</kbd> to start an inferior R process with arguments
3207 
<code>-d gdb</code>.
 3212 
<code>-d gdb</code>.
3208 
 
 3213 
 
3209 
   <p>A third option is to start an inferior R process via <small>ESS</small>
 3214 
   <p>A third option is to start an inferior R process via <small>ESS</small>
3210 
(<kbd>M-x R</kbd>) and then start GUD (<kbd>M-x gdb</kbd>) giving the R binary
 3215 
(<kbd>M-x R</kbd>) and then start GUD (<kbd>M-x gdb</kbd>) giving the R binary
3211 
(using its full path name) as the program to debug.  Use the program
 3216 
(using its full path name) as the program to debug.  Use the program
3212 
<code>ps</code> to find the process number of the currently running R
 3217 
<code>ps</code> to find the process number of the currently running R
3213 
process then use the <code>attach</code> command in gdb to attach it to that
 3218 
process then use the <code>attach</code> command in gdb to attach it to that
3214 
process.  One advantage of this method is that you have separate
 3219 
process.  One advantage of this method is that you have separate
3215 
<code>*R*</code> and <code>*gud-gdb*</code> windows.  Within the <code>*R*</code> window
 3220 
<code>*R*</code> and <code>*gud-gdb*</code> windows.  Within the <code>*R*</code> window
3216 
you have all the <small>ESS</small> facilities, such as object-name
 3221 
you have all the <small>ESS</small> facilities, such as object-name
3217 
completion, that we know and love.
 3222 
completion, that we know and love.
3218 
 
 3223 
 
3219 
   <p>When using GUD mode for debugging from within Emacs, you may find it
 3224 
   <p>When using GUD mode for debugging from within Emacs, you may find it
3220 
most convenient to use the directory with your code in it as the current
 3225 
most convenient to use the directory with your code in it as the current
3221 
working directory and then make a symbolic link from that directory to
 3226 
working directory and then make a symbolic link from that directory to
3222 
the R binary.  That way <code>.gdbinit</code> can stay in the directory with
 3227 
the R binary.  That way <code>.gdbinit</code> can stay in the directory with
3223 
the code and be used to set up the environment and the search paths for
 3228 
the code and be used to set up the environment and the search paths for
3224 
the source, e.g. as follows:
 3229 
the source, e.g. as follows:
3225 
 
 3230 
 
3226 
<pre class="example">     set env R_HOME /opt/R
 3231 
<pre class="example">     set env R_HOME /opt/R
3227 
     set env R_PAPERSIZE letter
 3232 
     set env R_PAPERSIZE letter
3228 
     set env R_PRINTCMD lpr
 3233 
     set env R_PRINTCMD lpr
3229 
     dir /opt/R/src/appl
 3234 
     dir /opt/R/src/appl
3230 
     dir /opt/R/src/main
 3235 
     dir /opt/R/src/main
3231 
     dir /opt/R/src/nmath
 3236 
     dir /opt/R/src/nmath
3232 
     dir /opt/R/src/unix
 3237 
     dir /opt/R/src/unix
3233 
     </pre>
 3238 
     </pre>
3234 
 
 3239 
 
3235 
<div class="node">
 3240 
<div class="node">
3236 
<p><hr>
 3241 
<p><hr>
3237 
Node:&nbsp;<a name="R%20Miscellanea">R Miscellanea</a>,
 3242 
Node:&nbsp;<a name="R%20Miscellanea">R Miscellanea</a>,
3238 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Programming">R Programming</a>,
 3243 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Programming">R Programming</a>,
3239 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20Emacs">R and Emacs</a>,
 3244 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20and%20Emacs">R and Emacs</a>,
3240 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 3245 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
3241 
<br>
 3246 
<br>
3242 
</div>
 3247 
</div>
3243 
 
 3248 
 
3244 
<h2 class="chapter">7 R Miscellanea</h2>
 3249 
<h2 class="chapter">7 R Miscellanea</h2>
3245 
 
 3250 
 
3246 
<ul class="menu">
 3251 
<ul class="menu">
3247 
<li><a accesskey="1" href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>:
3252 
<li><a accesskey="1" href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>:
3248 
<li><a accesskey="2" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>:
3253 
<li><a accesskey="2" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>:
3249 
<li><a accesskey="3" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>:
3254 
<li><a accesskey="3" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>:
3250 
<li><a accesskey="4" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>:
3255 
<li><a accesskey="4" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>:
3251 
<li><a accesskey="5" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>:
3256 
<li><a accesskey="5" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>:
3252 
<li><a accesskey="6" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>:
3257 
<li><a accesskey="6" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>:
3253 
<li><a accesskey="7" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>:
3258 
<li><a accesskey="7" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>:
3254 
<li><a accesskey="8" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>:
3259 
<li><a accesskey="8" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>:
3255 
<li><a accesskey="9" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>:
3260 
<li><a accesskey="9" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>:
3256 
<li><a href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>:
3261 
<li><a href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>:
3257 
<li><a href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>:
3262 
<li><a href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>:
3258 
<li><a href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>:
3263 
<li><a href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>:
3259 
<li><a href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>:
3264 
<li><a href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>:
3260 
<li><a href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>:
3265 
<li><a href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>:
3261 
<li><a href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>:
3266 
<li><a href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>:
3262 
<li><a href="#What%20are%20valid%20names%3f">What are valid names?</a>:
3267 
<li><a href="#What%20are%20valid%20names%3f">What are valid names?</a>:
3263 
<li><a href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>:
3268 
<li><a href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>:
3264 
<li><a href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>:
3269 
<li><a href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>:
3265 
<li><a href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>:
3270 
<li><a href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>:
3266 
<li><a href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>:
3271 
<li><a href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>:
3267 
<li><a href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>:
3272 
<li><a href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>:
3268 
<li><a href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>:
3273 
<li><a href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>:
3269 
<li><a href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>:
3274 
<li><a href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>:
3270 
<li><a href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>:
3275 
<li><a href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>:
3271 
<li><a href="#How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>:
3276 
<li><a href="#How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>:
3272 
</ul>
 3277 
</ul>
3273 
 
 3278 
 
3274 
<div class="node">
 3279 
<div class="node">
3275 
<p><hr>
 3280 
<p><hr>
3276 
Node:&nbsp;<a name="Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
 3281 
Node:&nbsp;<a name="Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
3277 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
 3282 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
3278 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Miscellanea">R Miscellanea</a>,
 3283 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Miscellanea">R Miscellanea</a>,
3279 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3284 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3280 
<br>
 3285 
<br>
3281 
</div>
 3286 
</div>
3282 
 
 3287 
 
3283 
<h3 class="section">7.1 Why does R run out of memory?</h3>
 3288 
<h3 class="section">7.1 Why does R run out of memory?</h3>
3284 
 
 3289 
 
3285 
<p>Versions of R prior to 1.2.0 used a <em>static</em> memory model.  At
 3290 
<p>Versions of R prior to 1.2.0 used a <em>static</em> memory model.  At
3286 
startup, R asked the operating system to reserve a fixed amount of
 3291 
startup, R asked the operating system to reserve a fixed amount of
3287 
memory for it.  The size of this chunk could not be changed
 3292 
memory for it.  The size of this chunk could not be changed
3288 
subsequently.  Hence, it could happen that not enough memory was
 3293 
subsequently.  Hence, it could happen that not enough memory was
3289 
allocated, e.g., when trying to read large data sets into R.  In such
 3294 
allocated, e.g., when trying to read large data sets into R.  In such
3290 
cases, it was necessary to restart R with more memory available, as
 3295 
cases, it was necessary to restart R with more memory available, as
3291 
controlled by the command line options <code>--nsize</code> and
 3296 
controlled by the command line options <code>--nsize</code> and
3292 
<code>--vsize</code>.
 3297 
<code>--vsize</code>.
3293 
 
 3298 
 
3294 
   <p>R version 1.2.0 introduces a new "generational" garbage collector,
 3299 
   <p>R version 1.2.0 introduces a new "generational" garbage collector,
3295 
which will increase the memory available to R as needed.  Hence, user
 3300 
which will increase the memory available to R as needed.  Hence, user
3296 
intervention is no longer necessary for ensuring that enough memory is
 3301 
intervention is no longer necessary for ensuring that enough memory is
3297 
available.
 3302 
available.
3298 
 
 3303 
 
3299 
   <p>The new garbage collector does not move objects in memory, meaning that
 3304 
   <p>The new garbage collector does not move objects in memory, meaning that
3300 
it is possible for the free memory to become fragmented so that large
 3305 
it is possible for the free memory to become fragmented so that large
3301 
objects cannot be allocated even when there is apparently enough memory
 3306 
objects cannot be allocated even when there is apparently enough memory
3302 
for them.
 3307 
for them.
3303 
 
 3308 
 
3304 
<div class="node">
 3309 
<div class="node">
3305 
<p><hr>
 3310 
<p><hr>
3306 
Node:&nbsp;<a name="Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
 3311 
Node:&nbsp;<a name="Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
3307 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
 3312 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
3308 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
 3313 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
3309 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3314 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3310 
<br>
 3315 
<br>
3311 
</div>
 3316 
</div>
3312 
 
 3317 
 
3313 
<h3 class="section">7.2 Why does sourcing a correct file fail?</h3>
 3318 
<h3 class="section">7.2 Why does sourcing a correct file fail?</h3>
3314 
 
 3319 
 
3315 
<p>Versions of R prior to 1.2.1 may have had problems parsing files not
 3320 
<p>Versions of R prior to 1.2.1 may have had problems parsing files not
3316 
ending in a newline.  Earlier R versions had a similar problem when
 3321 
ending in a newline.  Earlier R versions had a similar problem when
3317 
reading in data files.  This should no longer happen.
 3322 
reading in data files.  This should no longer happen.
3318 
 
 3323 
 
3319 
<div class="node">
 3324 
<div class="node">
3320 
<p><hr>
 3325 
<p><hr>
3321 
Node:&nbsp;<a name="How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
 3326 
Node:&nbsp;<a name="How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
3322 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
 3327 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
3323 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
 3328 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
3324 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3329 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3325 
<br>
 3330 
<br>
3326 
</div>
 3331 
</div>
3327 
 
 3332 
 
3328 
<h3 class="section">7.3 How can I set components of a list to NULL?</h3>
 3333 
<h3 class="section">7.3 How can I set components of a list to NULL?</h3>
3329 
 
 3334 
 
3330 
<p>You can use
 3335 
<p>You can use
3331 
 
 3336 
 
3332 
<pre class="example">     x[i] &lt;- list(NULL)
 3337 
<pre class="example">     x[i] &lt;- list(NULL)
3333 
     </pre>
 3338 
     </pre>
3334 
 
 3339 
 
3335 
<p>to set component <code>i</code> of the list <code>x</code> to <code>NULL</code>, similarly
 3340 
<p>to set component <code>i</code> of the list <code>x</code> to <code>NULL</code>, similarly
3336 
for named components.  Do not set <code>x[i]</code> or <code>x[[i]]</code> to
 3341 
for named components.  Do not set <code>x[i]</code> or <code>x[[i]]</code> to
3337 
<code>NULL</code>, because this will remove the corresponding component from
 3342 
<code>NULL</code>, because this will remove the corresponding component from
3338 
the list.
 3343 
the list.
3339 
 
 3344 
 
3340 
   <p>For dropping the row names of a matrix <code>x</code>, it may be easier to use
 3345 
   <p>For dropping the row names of a matrix <code>x</code>, it may be easier to use
3341 
<code>rownames(x) &lt;- NULL</code>, similarly for column names.
 3346 
<code>rownames(x) &lt;- NULL</code>, similarly for column names.
3342 
 
 3347 
 
3343 
<div class="node">
 3348 
<div class="node">
3344 
<p><hr>
 3349 
<p><hr>
3345 
Node:&nbsp;<a name="How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
 3350 
Node:&nbsp;<a name="How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
3346 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
 3351 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
3347 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
 3352 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
3348 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3353 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3349 
<br>
 3354 
<br>
3350 
</div>
 3355 
</div>
3351 
 
 3356 
 
3352 
<h3 class="section">7.4 How can I save my workspace?</h3>
 3357 
<h3 class="section">7.4 How can I save my workspace?</h3>
3353 
 
 3358 
 
3354 
<p><code>save.image()</code> saves the objects in the user's <code>.GlobalEnv</code> to
 3359 
<p><code>save.image()</code> saves the objects in the user's <code>.GlobalEnv</code> to
3355 
the file <code>.RData</code> in the R startup directory.  (This is also what
 3360 
the file <code>.RData</code> in the R startup directory.  (This is also what
3356 
happens after <kbd>q("yes")</kbd>.)  Using <code>save.image(</code><var>file</var><code>)</code> one
 3361 
happens after <kbd>q("yes")</kbd>.)  Using <code>save.image(</code><var>file</var><code>)</code> one
3357 
can save the image under a different name.
 3362 
can save the image under a different name.
3358 
 
 3363 
 
3359 
<div class="node">
 3364 
<div class="node">
3360 
<p><hr>
 3365 
<p><hr>
3361 
Node:&nbsp;<a name="How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
 3366 
Node:&nbsp;<a name="How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
3362 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
 3367 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
3363 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
 3368 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
3364 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3369 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3365 
<br>
 3370 
<br>
3366 
</div>
 3371 
</div>
3367 
 
 3372 
 
3368 
<h3 class="section">7.5 How can I clean up my workspace?</h3>
 3373 
<h3 class="section">7.5 How can I clean up my workspace?</h3>
3369 
 
 3374 
 
3370 
<p>To remove all objects in the currently active environment (typically
 3375 
<p>To remove all objects in the currently active environment (typically
3371 
<code>.GlobalEnv</code>), you can do
 3376 
<code>.GlobalEnv</code>), you can do
3372 
 
 3377 
 
3373 
<pre class="example">     rm(list = ls(all = TRUE))
 3378 
<pre class="example">     rm(list = ls(all = TRUE))
3374 
     </pre>
 3379 
     </pre>
3375 
 
 3380 
 
3376 
<p>(Without <code>all = TRUE</code>, only the objects with names not starting
 3381 
<p>(Without <code>all = TRUE</code>, only the objects with names not starting
3377 
with a <code>.</code> are removed.)
 3382 
with a <code>.</code> are removed.)
3378 
 
 3383 
 
3379 
<div class="node">
 3384 
<div class="node">
3380 
<p><hr>
 3385 
<p><hr>
3381 
Node:&nbsp;<a name="How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
 3386 
Node:&nbsp;<a name="How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
3382 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
 3387 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
3383 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
 3388 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
3384 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3389 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3385 
<br>
 3390 
<br>
3386 
</div>
 3391 
</div>
3387 
 
 3392 
 
3388 
<h3 class="section">7.6 How can I get eval() and D() to work?</h3>
 3393 
<h3 class="section">7.6 How can I get eval() and D() to work?</h3>
3389 
 
 3394 
 
3390 
<p>Strange things will happen if you use <code>eval(print(x), envir = e)</code>
 3395 
<p>Strange things will happen if you use <code>eval(print(x), envir = e)</code>
3391 
or <code>D(x^2, "x")</code>.  The first one will either tell you that
 3396 
or <code>D(x^2, "x")</code>.  The first one will either tell you that
3392 
"<code>x</code>" is not found, or print the value of the wrong <code>x</code>.
3397 
"<code>x</code>" is not found, or print the value of the wrong <code>x</code>.
3393 
The other one will likely return zero if <code>x</code> exists, and an error
 3398 
The other one will likely return zero if <code>x</code> exists, and an error
3394 
otherwise.
 3399 
otherwise.
3395 
 
 3400 
 
3396 
   <p>This is because in both cases, the first argument is evaluated in the
 3401 
   <p>This is because in both cases, the first argument is evaluated in the
3397 
calling environment first.  The result (which should be an object of
 3402 
calling environment first.  The result (which should be an object of
3398 
mode <code>"expression"</code> or <code>"call"</code>) is then evaluated or
 3403 
mode <code>"expression"</code> or <code>"call"</code>) is then evaluated or
3399 
differentiated.  What you (most likely) really want is obtained by
 3404 
differentiated.  What you (most likely) really want is obtained by
3400 
"quoting" the first argument upon surrounding it with
 3405 
"quoting" the first argument upon surrounding it with
3401 
<code>expression()</code>.  For example,
 3406 
<code>expression()</code>.  For example,
3402 
 
 3407 
 
3403 
<pre class="example">     R&gt; D(expression(x^2), "x")
 3408 
<pre class="example">     R&gt; D(expression(x^2), "x")
3404 
     2 * x
 3409 
     2 * x
3405 
     </pre>
 3410 
     </pre>
3406 
 
 3411 
 
3407 
   <p>Although this behavior may initially seem to be rather strange, is
 3412 
   <p>Although this behavior may initially seem to be rather strange, is
3408 
perfectly logical.  The "intuitive" behavior could easily be
 3413 
perfectly logical.  The "intuitive" behavior could easily be
3409 
implemented, but problems would arise whenever the expression is
 3414 
implemented, but problems would arise whenever the expression is
3410 
contained in a variable, passed as a parameter, or is the result of a
 3415 
contained in a variable, passed as a parameter, or is the result of a
3411 
function call.  Consider for instance the semantics in cases like
 3416 
function call.  Consider for instance the semantics in cases like
3412 
 
 3417 
 
3413 
<pre class="example">     D2 &lt;- function(e, n) D(D(e, n), n)
 3418 
<pre class="example">     D2 &lt;- function(e, n) D(D(e, n), n)
3414 
     </pre>
 3419 
     </pre>
3415 
 
 3420 
 
3416 
<p>or
 3421 
<p>or
3417 
 
 3422 
 
3418 
<pre class="example">     g &lt;- function(y) eval(substitute(y), sys.frame(sys.parent(n = 2)))
 3423 
<pre class="example">     g &lt;- function(y) eval(substitute(y), sys.frame(sys.parent(n = 2)))
3419 
     g(a * b)
 3424 
     g(a * b)
3420 
     </pre>
 3425 
     </pre>
3421 
 
 3426 
 
3422 
   <p>See the help page for <code>deriv()</code> for more examples.
 3427 
   <p>See the help page for <code>deriv()</code> for more examples.
3423 
 
 3428 
 
3424 
<div class="node">
 3429 
<div class="node">
3425 
<p><hr>
 3430 
<p><hr>
3426 
Node:&nbsp;<a name="Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
 3431 
Node:&nbsp;<a name="Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
3427 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
 3432 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
3428 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
 3433 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
3429 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3434 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3430 
<br>
 3435 
<br>
3431 
</div>
 3436 
</div>
3432 
 
 3437 
 
3433 
<h3 class="section">7.7 Why do my matrices lose dimensions?</h3>
 3438 
<h3 class="section">7.7 Why do my matrices lose dimensions?</h3>
3434 
 
 3439 
 
3435 
<p>When a matrix with a single row or column is created by a subscripting
 3440 
<p>When a matrix with a single row or column is created by a subscripting
3436 
operation, e.g., <code>row &lt;- mat[2, ]</code>, it is by default turned into a
 3441 
operation, e.g., <code>row &lt;- mat[2, ]</code>, it is by default turned into a
3437 
vector.  In a similar way if an array with dimension, say, 2&nbsp;x&nbsp;3&nbsp;x&nbsp;1&nbsp;x&nbsp;4 is created by subscripting it will be coerced into a 2&nbsp;x&nbsp;3&nbsp;x&nbsp;4
 3442 
vector.  In a similar way if an array with dimension, say, 2&nbsp;x&nbsp;3&nbsp;x&nbsp;1&nbsp;x&nbsp;4 is created by subscripting it will be coerced into a 2&nbsp;x&nbsp;3&nbsp;x&nbsp;4
3438 
array, losing the unnecessary dimension.  After much discussion this has
 3443 
array, losing the unnecessary dimension.  After much discussion this has
3439 
been determined to be a <em>feature</em>.
 3444 
been determined to be a <em>feature</em>.
3440 
 
 3445 
 
3441 
   <p>To prevent this happening, add the option <code>drop = FALSE</code> to the
 3446 
   <p>To prevent this happening, add the option <code>drop = FALSE</code> to the
3442 
subscripting.  For example,
 3447 
subscripting.  For example,
3443 
 
 3448 
 
3444 
<pre class="example">     rowmatrix &lt;- mat[2, , drop = FALSE]  # creates a row matrix
 3449 
<pre class="example">     rowmatrix &lt;- mat[2, , drop = FALSE]  # creates a row matrix
3445 
     colmatrix &lt;- mat[, 2, drop = FALSE]  # creates a column matrix
 3450 
     colmatrix &lt;- mat[, 2, drop = FALSE]  # creates a column matrix
3446 
     a &lt;- b[1, 1, 1, drop = FALSE]        # creates a 1 x 1 x 1 array
 3451 
     a &lt;- b[1, 1, 1, drop = FALSE]        # creates a 1 x 1 x 1 array
3447 
     </pre>
 3452 
     </pre>
3448 
 
 3453 
 
3449 
   <p>The <code>drop = FALSE</code> option should be used defensively when
 3454 
   <p>The <code>drop = FALSE</code> option should be used defensively when
3450 
programming.  For example, the statement
 3455 
programming.  For example, the statement
3451 
 
 3456 
 
3452 
<pre class="example">     somerows &lt;- mat[index, ]
 3457 
<pre class="example">     somerows &lt;- mat[index, ]
3453 
     </pre>
 3458 
     </pre>
3454 
 
 3459 
 
3455 
<p>will return a vector rather than a matrix if <code>index</code> happens to
 3460 
<p>will return a vector rather than a matrix if <code>index</code> happens to
3456 
have length 1, causing errors later in the code.  It should probably be
 3461 
have length 1, causing errors later in the code.  It should probably be
3457 
rewritten as
 3462 
rewritten as
3458 
 
 3463 
 
3459 
<pre class="example">     somerows &lt;- mat[index, , drop = FALSE]
 3464 
<pre class="example">     somerows &lt;- mat[index, , drop = FALSE]
3460 
     </pre>
 3465 
     </pre>
3461 
 
 3466 
 
3462 
<div class="node">
 3467 
<div class="node">
3463 
<p><hr>
 3468 
<p><hr>
3464 
Node:&nbsp;<a name="How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
 3469 
Node:&nbsp;<a name="How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
3465 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
 3470 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
3466 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
 3471 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
3467 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3472 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3468 
<br>
 3473 
<br>
3469 
</div>
 3474 
</div>
3470 
 
 3475 
 
3471 
<h3 class="section">7.8 How does autoloading work?</h3>
 3476 
<h3 class="section">7.8 How does autoloading work?</h3>
3472 
 
 3477 
 
3473 
<p>R has a special environment called <code>.AutoloadEnv</code>.  Using
 3478 
<p>R has a special environment called <code>.AutoloadEnv</code>.  Using
3474 
<kbd>autoload(</kbd><var>name</var><kbd>, </kbd><var>pkg</var><kbd>)</kbd>, where <var>name</var> and
 3479 
<kbd>autoload(</kbd><var>name</var><kbd>, </kbd><var>pkg</var><kbd>)</kbd>, where <var>name</var> and
3475 
<var>pkg</var> are strings giving the names of an object and the package
 3480 
<var>pkg</var> are strings giving the names of an object and the package
3476 
containing it, stores some information in this environment.  When R
 3481 
containing it, stores some information in this environment.  When R
3477 
tries to evaluate <var>name</var>, it loads the corresponding package
 3482 
tries to evaluate <var>name</var>, it loads the corresponding package
3478 
<var>pkg</var> and reevaluates <var>name</var> in the new package's
 3483 
<var>pkg</var> and reevaluates <var>name</var> in the new package's
3479 
environment.
 3484 
environment.
3480 
 
 3485 
 
3481 
   <p>Using this mechanism makes R behave as if the package was loaded, but
 3486 
   <p>Using this mechanism makes R behave as if the package was loaded, but
3482 
does not occupy memory (yet).
 3487 
does not occupy memory (yet).
3483 
 
 3488 
 
3484 
   <p>See the help page for <code>autoload()</code> for a very nice example.
 3489 
   <p>See the help page for <code>autoload()</code> for a very nice example.
3485 
 
 3490 
 
3486 
<div class="node">
 3491 
<div class="node">
3487 
<p><hr>
 3492 
<p><hr>
3488 
Node:&nbsp;<a name="How%20should%20I%20set%20options%3f">How should I set options?</a>,
 3493 
Node:&nbsp;<a name="How%20should%20I%20set%20options%3f">How should I set options?</a>,
3489 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
 3494 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
3490 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
 3495 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
3491 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3496 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3492 
<br>
 3497 
<br>
3493 
</div>
 3498 
</div>
3494 
 
 3499 
 
3495 
<h3 class="section">7.9 How should I set options?</h3>
 3500 
<h3 class="section">7.9 How should I set options?</h3>
3496 
 
 3501 
 
3497 
<p>The function <code>options()</code> allows setting and examining a variety of
 3502 
<p>The function <code>options()</code> allows setting and examining a variety of
3498 
global "options" which affect the way in which R computes and displays
 3503 
global "options" which affect the way in which R computes and displays
3499 
its results.  The variable <code>.Options</code> holds the current values of
 3504 
its results.  The variable <code>.Options</code> holds the current values of
3500 
these options, but should never directly be assigned to unless you want
 3505 
these options, but should never directly be assigned to unless you want
3501 
to drive yourself crazy--simply pretend that it is a "read-only"
 3506 
to drive yourself crazy--simply pretend that it is a "read-only"
3502 
variable.
 3507 
variable.
3503 
 
 3508 
 
3504 
   <p>For example, given
 3509 
   <p>For example, given
3505 
 
 3510 
 
3506 
<pre class="example">     test1 &lt;- function(x = pi, dig = 3) {
 3511 
<pre class="example">     test1 &lt;- function(x = pi, dig = 3) {
3507 
       oo &lt;- options(digits = dig); on.exit(options(oo));
 3512 
       oo &lt;- options(digits = dig); on.exit(options(oo));
3508 
       cat(.Options$digits, x, "\n")
 3513 
       cat(.Options$digits, x, "\n")
3509 
     }
 3514 
     }
3510 
     test2 &lt;- function(x = pi, dig = 3) {
 3515 
     test2 &lt;- function(x = pi, dig = 3) {
3511 
       .Options$digits &lt;- dig
 3516 
       .Options$digits &lt;- dig
3512 
       cat(.Options$digits, x, "\n")
 3517 
       cat(.Options$digits, x, "\n")
3513 
     }
 3518 
     }
3514 
     </pre>
 3519 
     </pre>
3515 
 
 3520 
 
3516 
<p>we obtain:
 3521 
<p>we obtain:
3517 
 
 3522 
 
3518 
<pre class="example">     R&gt; test1()
 3523 
<pre class="example">     R&gt; test1()
3519 
     3 3.14
 3524 
     3 3.14
3520 
     R&gt; test2()
 3525 
     R&gt; test2()
3521 
     3 3.141593
 3526 
     3 3.141593
3522 
     </pre>
 3527 
     </pre>
3523 
 
 3528 
 
3524 
   <p>What is really used is the <em>global</em> value of <code>.Options</code>, and
 3529 
   <p>What is really used is the <em>global</em> value of <code>.Options</code>, and
3525 
using <kbd>options(OPT = VAL)</kbd> correctly updates it.  Local copies of
 3530 
using <kbd>options(OPT = VAL)</kbd> correctly updates it.  Local copies of
3526 
<code>.Options</code>, either in <code>.GlobalEnv</code> or in a function
 3531 
<code>.Options</code>, either in <code>.GlobalEnv</code> or in a function
3527 
environment (frame), are just silently disregarded.
 3532 
environment (frame), are just silently disregarded.
3528 
 
 3533 
 
3529 
<div class="node">
 3534 
<div class="node">
3530 
<p><hr>
 3535 
<p><hr>
3531 
Node:&nbsp;<a name="How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
 3536 
Node:&nbsp;<a name="How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
3532 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
 3537 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
3533 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
 3538 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
3534 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3539 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3535 
<br>
 3540 
<br>
3536 
</div>
 3541 
</div>
3537 
 
 3542 
 
3538 
<h3 class="section">7.10 How do file names work in Windows?</h3>
 3543 
<h3 class="section">7.10 How do file names work in Windows?</h3>
3539 
 
 3544 
 
3540 
<p>As R uses C-style string handling, <code>\</code> is treated as an escape
 3545 
<p>As R uses C-style string handling, <code>\</code> is treated as an escape
3541 
character, so that for example one can enter a newline as <code>\n</code>.
3546 
character, so that for example one can enter a newline as <code>\n</code>.
3542 
When you really need a <code>\</code>, you have to escape it with another
 3547 
When you really need a <code>\</code>, you have to escape it with another
3543 
<code>\</code>.
 3548 
<code>\</code>.
3544 
 
 3549 
 
3545 
   <p>Thus, in filenames use something like <code>"c:\\data\\money.dat"</code>.  You
 3550 
   <p>Thus, in filenames use something like <code>"c:\\data\\money.dat"</code>.  You
3546 
can also replace <code>\</code> by <code>/</code> (<code>"c:/data/money.dat"</code>).
 3551 
can also replace <code>\</code> by <code>/</code> (<code>"c:/data/money.dat"</code>).
3547 
 
 3552 
 
3548 
<div class="node">
 3553 
<div class="node">
3549 
<p><hr>
 3554 
<p><hr>
3550 
Node:&nbsp;<a name="Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
 3555 
Node:&nbsp;<a name="Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
3551 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
 3556 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
3552 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
 3557 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
3553 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3558 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3554 
<br>
 3559 
<br>
3555 
</div>
 3560 
</div>
3556 
 
 3561 
 
3557 
<h3 class="section">7.11 Why does plotting give a color allocation error?</h3>
 3562 
<h3 class="section">7.11 Why does plotting give a color allocation error?</h3>
3558 
 
 3563 
 
3559 
<p>Sometimes plotting, e.g., when running <code>demo("image")</code>, results in
 3564 
<p>Sometimes plotting, e.g., when running <code>demo("image")</code>, results in
3560 
"Error: color allocation error".  This is an X problem, and only
 3565 
"Error: color allocation error".  This is an X problem, and only
3561 
indirectly related to R.  It occurs when applications started prior to R
 3566 
indirectly related to R.  It occurs when applications started prior to R
3562 
have used all the available colors.  (How many colors are available
 3567 
have used all the available colors.  (How many colors are available
3563 
depends on the X configuration; sometimes only 256 colors can be used.)
 3568 
depends on the X configuration; sometimes only 256 colors can be used.)
3564 
 
 3569 
 
3565 
   <p>One application which is notorious for "eating" colors is Netscape.
3570 
   <p>One application which is notorious for "eating" colors is Netscape.
3566 
If the problem occurs when Netscape is running, try (re)starting it with
 3571 
If the problem occurs when Netscape is running, try (re)starting it with
3567 
either the <code>-no-install</code> (to use the default colormap) or the
 3572 
either the <code>-no-install</code> (to use the default colormap) or the
3568 
<code>-install</code> (to install a private colormap) option.
 3573 
<code>-install</code> (to install a private colormap) option.
3569 
 
 3574 
 
3570 
   <p>You could also set the <code>colortype</code> of <code>X11()</code> to
 3575 
   <p>You could also set the <code>colortype</code> of <code>X11()</code> to
3571 
<code>"pseudo.cube"</code> rather than the default <code>"pseudo"</code>.  See the
 3576 
<code>"pseudo.cube"</code> rather than the default <code>"pseudo"</code>.  See the
3572 
help page for <code>X11()</code> for more information.
 3577 
help page for <code>X11()</code> for more information.
3573 
 
 3578 
 
3574 
<div class="node">
 3579 
<div class="node">
3575 
<p><hr>
 3580 
<p><hr>
3576 
Node:&nbsp;<a name="How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
 3581 
Node:&nbsp;<a name="How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
3577 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
 3582 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
3578 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
 3583 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
3579 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3584 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3580 
<br>
 3585 
<br>
3581 
</div>
 3586 
</div>
3582 
 
 3587 
 
3583 
<h3 class="section">7.12 How do I convert factors to numeric?</h3>
 3588 
<h3 class="section">7.12 How do I convert factors to numeric?</h3>
3584 
 
 3589 
 
3585 
<p>It may happen that when reading numeric data into R (usually, when
 3590 
<p>It may happen that when reading numeric data into R (usually, when
3586 
reading in a file), they come in as factors.  If <code>f</code> is such a
 3591 
reading in a file), they come in as factors.  If <code>f</code> is such a
3587 
factor object, you can use
 3592 
factor object, you can use
3588 
 
 3593 
 
3589 
<pre class="example">     as.numeric(as.character(f))
 3594 
<pre class="example">     as.numeric(as.character(f))
3590 
     </pre>
 3595 
     </pre>
3591 
 
 3596 
 
3592 
<p>to get the numbers back.  More efficient, but harder to remember, is
 3597 
<p>to get the numbers back.  More efficient, but harder to remember, is
3593 
 
 3598 
 
3594 
<pre class="example">     as.numeric(levels(f))[as.integer(f)]
 3599 
<pre class="example">     as.numeric(levels(f))[as.integer(f)]
3595 
     </pre>
 3600 
     </pre>
3596 
 
 3601 
 
3597 
   <p>In any case, do not call <code>as.numeric()</code> or their likes directly for
 3602 
   <p>In any case, do not call <code>as.numeric()</code> or their likes directly for
3598 
the task at hand (as <code>as.numeric()</code> or <code>unclass()</code> give the
 3603 
the task at hand (as <code>as.numeric()</code> or <code>unclass()</code> give the
3599 
internal codes).
 3604 
internal codes).
3600 
 
 3605 
 
3601 
<div class="node">
 3606 
<div class="node">
3602 
<p><hr>
 3607 
<p><hr>
3603 
Node:&nbsp;<a name="Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
 3608 
Node:&nbsp;<a name="Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
3604 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
 3609 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
3605 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
 3610 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20I%20convert%20factors%20to%20numeric%3f">How do I convert factors to numeric?</a>,
3606 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3611 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3607 
<br>
 3612 
<br>
3608 
</div>
 3613 
</div>
3609 
 
 3614 
 
3610 
<h3 class="section">7.13 Are Trellis displays implemented in R?</h3>
 3615 
<h3 class="section">7.13 Are Trellis displays implemented in R?</h3>
3611 
 
 3616 
 
3612 
<p>The recommended package <strong>lattice</strong> (which is based on another
 3617 
<p>The recommended package <strong>lattice</strong> (which is based on another
3613 
recommended package, <strong>grid</strong>) provides graphical functionality
 3618 
recommended package, <strong>grid</strong>) provides graphical functionality
3614 
that is compatible with most Trellis commands.
 3619 
that is compatible with most Trellis commands.
3615 
 
 3620 
 
3616 
   <p>You could also look at <code>coplot()</code> and <code>dotchart()</code> which might
 3621 
   <p>You could also look at <code>coplot()</code> and <code>dotchart()</code> which might
3617 
do at least some of what you want.  Note also that the R version of
 3622 
do at least some of what you want.  Note also that the R version of
3618 
<code>pairs()</code> is fairly general and provides most of the functionality
 3623 
<code>pairs()</code> is fairly general and provides most of the functionality
3619 
of <code>splom()</code>, and that R's default plot method has an argument
 3624 
of <code>splom()</code>, and that R's default plot method has an argument
3620 
<code>asp</code> allowing to specify (and fix against device resizing) the
 3625 
<code>asp</code> allowing to specify (and fix against device resizing) the
3621 
aspect ratio of the plot.
 3626 
aspect ratio of the plot.
3622 
 
 3627 
 
3623 
   <p>(Because the word "Trellis" has been claimed as a trademark we do not
 3628 
   <p>(Because the word "Trellis" has been claimed as a trademark we do not
3624 
use it in R.  The name "lattice" has been chosen for the R
 3629 
use it in R.  The name "lattice" has been chosen for the R
3625 
equivalent.)
 3630 
equivalent.)
3626 
 
 3631 
 
3627 
<div class="node">
 3632 
<div class="node">
3628 
<p><hr>
 3633 
<p><hr>
3629 
Node:&nbsp;<a name="What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
 3634 
Node:&nbsp;<a name="What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
3630 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
 3635 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
3631 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
 3636 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20Trellis%20displays%20implemented%20in%20R%3f">Are Trellis displays implemented in R?</a>,
3632 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3637 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3633 
<br>
 3638 
<br>
3634 
</div>
 3639 
</div>
3635 
 
 3640 
 
3636 
<h3 class="section">7.14 What are the enclosing and parent environments?</h3>
 3641 
<h3 class="section">7.14 What are the enclosing and parent environments?</h3>
3637 
 
 3642 
 
3638 
<p>Inside a function you may want to access variables in two additional
 3643 
<p>Inside a function you may want to access variables in two additional
3639 
environments: the one that the function was defined in ("enclosing"),
 3644 
environments: the one that the function was defined in ("enclosing"),
3640 
and the one it was invoked in ("parent").
 3645 
and the one it was invoked in ("parent").
3641 
 
 3646 
 
3642 
   <p>If you create a function at the command line or load it in a package its
 3647 
   <p>If you create a function at the command line or load it in a package its
3643 
enclosing environment is the global workspace.  If you define a function
 3648 
enclosing environment is the global workspace.  If you define a function
3644 
<code>f()</code> inside another function <code>g()</code> its enclosing environment
 3649 
<code>f()</code> inside another function <code>g()</code> its enclosing environment
3645 
is the environment inside <code>g()</code>.  The enclosing environment for a
 3650 
is the environment inside <code>g()</code>.  The enclosing environment for a
3646 
function is fixed when the function is created.  You can find out the
 3651 
function is fixed when the function is created.  You can find out the
3647 
enclosing environment for a function <code>f()</code> using
 3652 
enclosing environment for a function <code>f()</code> using
3648 
<code>environment(f)</code>.
 3653 
<code>environment(f)</code>.
3649 
 
 3654 
 
3650 
   <p>The "parent" environment, on the other hand, is defined when you
 3655 
   <p>The "parent" environment, on the other hand, is defined when you
3651 
invoke a function.  If you invoke <code>lm()</code> at the command line its
 3656 
invoke a function.  If you invoke <code>lm()</code> at the command line its
3652 
parent environment is the global workspace, if you invoke it inside a
 3657 
parent environment is the global workspace, if you invoke it inside a
3653 
function <code>f()</code> then its parent environment is the environment
 3658 
function <code>f()</code> then its parent environment is the environment
3654 
inside <code>f()</code>.  You can find out the parent environment for an
 3659 
inside <code>f()</code>.  You can find out the parent environment for an
3655 
invocation of a function by using <code>parent.frame()</code> or
 3660 
invocation of a function by using <code>parent.frame()</code> or
3656 
<code>sys.frame(sys.parent())</code>.
 3661 
<code>sys.frame(sys.parent())</code>.
3657 
 
 3662 
 
3658 
   <p>So for most user-visible functions the enclosing environment will be the
 3663 
   <p>So for most user-visible functions the enclosing environment will be the
3659 
global workspace, since that is where most functions are defined.  The
 3664 
global workspace, since that is where most functions are defined.  The
3660 
parent environment will be wherever the function happens to be called
 3665 
parent environment will be wherever the function happens to be called
3661 
from.  If a function <code>f()</code> is defined inside another function
 3666 
from.  If a function <code>f()</code> is defined inside another function
3662 
<code>g()</code> it will probably be used inside <code>g()</code> as well, so its
 3667 
<code>g()</code> it will probably be used inside <code>g()</code> as well, so its
3663 
parent environment and enclosing environment will probably be the same.
 3668 
parent environment and enclosing environment will probably be the same.
3664 
 
 3669 
 
3665 
   <p>Parent environments are important because things like model formulas
 3670 
   <p>Parent environments are important because things like model formulas
3666 
need to be evaluated in the environment the function was called from,
 3671 
need to be evaluated in the environment the function was called from,
3667 
since that's where all the variables will be available.  This relies on
 3672 
since that's where all the variables will be available.  This relies on
3668 
the parent environment being potentially different with each invocation.
 3673 
the parent environment being potentially different with each invocation.
3669 
 
 3674 
 
3670 
   <p>Enclosing environments are important because a function can use
 3675 
   <p>Enclosing environments are important because a function can use
3671 
variables in the enclosing environment to share information with other
 3676 
variables in the enclosing environment to share information with other
3672 
functions or with other invocations of itself (see the section on
 3677 
functions or with other invocations of itself (see the section on
3673 
lexical scoping).  This relies on the enclosing environment being the
 3678 
lexical scoping).  This relies on the enclosing environment being the
3674 
same each time the function is invoked.
 3679 
same each time the function is invoked.
3675 
 
 3680 
 
3676 
   <p>Scoping <em>is</em> hard.  Looking at examples helps.  It is particularly
 3681 
   <p>Scoping <em>is</em> hard.  Looking at examples helps.  It is particularly
3677 
instructive to look at examples that work differently in R and S and try
 3682 
instructive to look at examples that work differently in R and S and try
3678 
to see why they differ.  One way to describe the scoping differences
 3683 
to see why they differ.  One way to describe the scoping differences
3679 
between R and S is to say that in S the enclosing environment is
 3684 
between R and S is to say that in S the enclosing environment is
3680 
<em>always</em> the global workspace, but in R the enclosing environment
 3685 
<em>always</em> the global workspace, but in R the enclosing environment
3681 
is wherever the function was created.
 3686 
is wherever the function was created.
3682 
 
 3687 
 
3683 
<div class="node">
 3688 
<div class="node">
3684 
<p><hr>
 3689 
<p><hr>
3685 
Node:&nbsp;<a name="How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
 3690 
Node:&nbsp;<a name="How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
3686 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20valid%20names%3f">What are valid names?</a>,
 3691 
Next:&nbsp;<a rel="next" accesskey="n" href="#What%20are%20valid%20names%3f">What are valid names?</a>,
3687 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
 3692 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20the%20enclosing%20and%20parent%20environments%3f">What are the enclosing and parent environments?</a>,
3688 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3693 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3689 
<br>
 3694 
<br>
3690 
</div>
 3695 
</div>
3691 
 
 3696 
 
3692 
<h3 class="section">7.15 How can I substitute into a plot label?</h3>
 3697 
<h3 class="section">7.15 How can I substitute into a plot label?</h3>
3693 
 
 3698 
 
3694 
<p>Often, it is desired to use the value of an R object in a plot label,
 3699 
<p>Often, it is desired to use the value of an R object in a plot label,
3695 
e.g., a title.  This is easily accomplished using <code>paste()</code> if the
 3700 
e.g., a title.  This is easily accomplished using <code>paste()</code> if the
3696 
label is a simple character string, but not always obvious in case the
 3701 
label is a simple character string, but not always obvious in case the
3697 
label is an expression (for refined mathematical annotation).  In such a
 3702 
label is an expression (for refined mathematical annotation).  In such a
3698 
case, either use <code>parse()</code> on your pasted character string or use
 3703 
case, either use <code>parse()</code> on your pasted character string or use
3699 
<code>substitute()</code> on an expression.  For example, if <code>ahat</code> is an
 3704 
<code>substitute()</code> on an expression.  For example, if <code>ahat</code> is an
3700 
estimator of your parameter a of interest, use
 3705 
estimator of your parameter a of interest, use
3701 
 
 3706 
 
3702 
<pre class="example">     title(substitute(hat(a) == ahat, list(ahat = ahat)))
 3707 
<pre class="example">     title(substitute(hat(a) == ahat, list(ahat = ahat)))
3703 
     </pre>
 3708 
     </pre>
3704 
 
 3709 
 
3705 
<p>(note that it is <code>==</code> and not <code>=</code>).  There are more worked
 3710 
<p>(note that it is <code>==</code> and not <code>=</code>).  There are more worked
3706 
examples in the mailing list achives.
 3711 
examples in the mailing list achives.
3707 
 
 3712 
 
3708 
<div class="node">
 3713 
<div class="node">
3709 
<p><hr>
 3714 
<p><hr>
3710 
Node:&nbsp;<a name="What%20are%20valid%20names%3f">What are valid names?</a>,
 3715 
Node:&nbsp;<a name="What%20are%20valid%20names%3f">What are valid names?</a>,
3711 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
 3716 
Next:&nbsp;<a rel="next" accesskey="n" href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
3712 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
 3717 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20substitute%20into%20a%20plot%20label%3f">How can I substitute into a plot label?</a>,
3713 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3718 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3714 
<br>
 3719 
<br>
3715 
</div>
 3720 
</div>
3716 
 
 3721 
 
3717 
<h3 class="section">7.16 What are valid names?</h3>
 3722 
<h3 class="section">7.16 What are valid names?</h3>
3718 
 
 3723 
 
3719 
<p>When creating data frames using <code>data.frame()</code> or
 3724 
<p>When creating data frames using <code>data.frame()</code> or
3720 
<code>read.table()</code>, R by default ensures that the variable names are
 3725 
<code>read.table()</code>, R by default ensures that the variable names are
3721 
syntactically valid.  (The argument <code>check.names</code> to these
 3726 
syntactically valid.  (The argument <code>check.names</code> to these
3722 
functions controls whether variable names are checked and adjusted by
 3727 
functions controls whether variable names are checked and adjusted by
3723 
<code>make.names()</code> if needed.)
 3728 
<code>make.names()</code> if needed.)
3724 
 
 3729 
 
3725 
   <p>To understand what names are "valid", one needs to take into account
 3730 
   <p>To understand what names are "valid", one needs to take into account
3726 
that the term "name" is used in several different (but related) ways
 3731 
that the term "name" is used in several different (but related) ways
3727 
in the language:
 3732 
in the language:
3728 
 
 3733 
 
3729 
     <ol type=1 start=1>
 3734 
     <ol type=1 start=1>
3730 
<li>A <em>syntactic name</em> is a string the parser interprets as this type
 3735 
<li>A <em>syntactic name</em> is a string the parser interprets as this type
3731 
of expression.  It consists of letters, numbers, and the dot and (for
 3736 
of expression.  It consists of letters, numbers, and the dot and (for
3732 
version of R at least 1.9.0) underscore characters, and starts with
 3737 
version of R at least 1.9.0) underscore characters, and starts with
3733 
either a letter or a dot not followed by a number.  Reserved words are
 3738 
either a letter or a dot not followed by a number.  Reserved words are
3734 
not syntactic names.
3739 
not syntactic names.
3735 
<li>An <em>object name</em> is a string associated with an object that is
 3740 
<li>An <em>object name</em> is a string associated with an object that is
3736 
assigned in an expression either by having the object name on the left
 3741 
assigned in an expression either by having the object name on the left
3737 
of an assignment operation or as an argument to the <code>assign()</code>
 3742 
of an assignment operation or as an argument to the <code>assign()</code>
3738 
function.  It is usually a syntactic name as well, but can be any
 3743 
function.  It is usually a syntactic name as well, but can be any
3739 
non-empty string if it is quoted (and it is always quoted in the call to
 3744 
non-empty string if it is quoted (and it is always quoted in the call to
3740 
<code>assign()</code>).
 3745 
<code>assign()</code>).
3741 
 
 3746 
 
3742 
     <li>An <em>argument name</em> is what appears to the left of the equals sign
 3747 
     <li>An <em>argument name</em> is what appears to the left of the equals sign
3743 
when supplying an argument in a function call (for example,
 3748 
when supplying an argument in a function call (for example,
3744 
<code>f(trim=.5)</code>).  Argument names are also usually syntactic names,
 3749 
<code>f(trim=.5)</code>).  Argument names are also usually syntactic names,
3745 
but again can be anything if they are quoted.
 3750 
but again can be anything if they are quoted.
3746 
 
 3751 
 
3747 
     <li>An <em>element name</em> is a string that identifies a piece of an object
 3752 
     <li>An <em>element name</em> is a string that identifies a piece of an object
3748 
(a component of a list, for example.)  When it is used on the right of
 3753 
(a component of a list, for example.)  When it is used on the right of
3749 
the <code>$</code> operator, it must be a syntactic name, or quoted.
3754 
the <code>$</code> operator, it must be a syntactic name, or quoted.
3750 
Otherwise, element names can be any strings.  (When an object is used as
 3755 
Otherwise, element names can be any strings.  (When an object is used as
3751 
a database, as in a call to <code>eval()</code> or <code>attach()</code>, the
 3756 
a database, as in a call to <code>eval()</code> or <code>attach()</code>, the
3752 
element names become object names.)
 3757 
element names become object names.)
3753 
 
 3758 
 
3754 
     <li>Finally, a <em>file name</em> is a string identifying a file in the
 3759 
     <li>Finally, a <em>file name</em> is a string identifying a file in the
3755 
operating system for reading, writing, etc.  It really has nothing much
 3760 
operating system for reading, writing, etc.  It really has nothing much
3756 
to do with names in the language, but it is traditional to call these
 3761 
to do with names in the language, but it is traditional to call these
3757 
strings file "names".
 3762 
strings file "names".
3758 
        </ol>
 3763 
        </ol>
3759 
 
 3764 
 
3760 
<div class="node">
 3765 
<div class="node">
3761 
<p><hr>
 3766 
<p><hr>
3762 
Node:&nbsp;<a name="Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
 3767 
Node:&nbsp;<a name="Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
3763 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
 3768 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
3764 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20valid%20names%3f">What are valid names?</a>,
 3769 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20are%20valid%20names%3f">What are valid names?</a>,
3765 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3770 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3766 
<br>
 3771 
<br>
3767 
</div>
 3772 
</div>
3768 
 
 3773 
 
3769 
<h3 class="section">7.17 Are GAMs implemented in R?</h3>
 3774 
<h3 class="section">7.17 Are GAMs implemented in R?</h3>
3770 
 
 3775 
 
3771 
<p>There is a <code>gam()</code> function for Generalized Additive Models in
 3776 
<p>There is a <code>gam()</code> function for Generalized Additive Models in
3772 
package <strong>mgcv</strong>, but it is not an exact clone of what is described
 3777 
package <strong>mgcv</strong>, but it is not an exact clone of what is described
3773 
in the White Book (no <code>lo()</code> for example).  Package <strong>gss</strong>
 3778 
in the White Book (no <code>lo()</code> for example).  Package <strong>gss</strong>
3774 
can fit spline-based GAMs too.  And if you can accept regression splines
 3779 
can fit spline-based GAMs too.  And if you can accept regression splines
3775 
you can use <code>glm()</code>.  For gaussian GAMs you can use <code>bruto()</code>
 3780 
you can use <code>glm()</code>.  For gaussian GAMs you can use <code>bruto()</code>
3776 
from package <strong>mda</strong>.
 3781 
from package <strong>mda</strong>.
3777 
 
 3782 
 
3778 
<div class="node">
 3783 
<div class="node">
3779 
<p><hr>
 3784 
<p><hr>
3780 
Node:&nbsp;<a name="Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
 3785 
Node:&nbsp;<a name="Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
3781 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
 3786 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
3782 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
 3787 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Are%20GAMs%20implemented%20in%20R%3f">Are GAMs implemented in R?</a>,
3783 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3788 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3784 
<br>
 3789 
<br>
3785 
</div>
 3790 
</div>
3786 
 
 3791 
 
3787 
<h3 class="section">7.18 Why is the output not printed when I source() a file?</h3>
 3792 
<h3 class="section">7.18 Why is the output not printed when I source() a file?</h3>
3788 
 
 3793 
 
3789 
<p>Most R commands do not generate any output. The command
 3794 
<p>Most R commands do not generate any output. The command
3790 
 
 3795 
 
3791 
<pre class="example">     1+1
 3796 
<pre class="example">     1+1
3792 
     </pre>
 3797 
     </pre>
3793 
 
 3798 
 
3794 
<p>computes the value 2 and returns it; the command
 3799 
<p>computes the value 2 and returns it; the command
3795 
 
 3800 
 
3796 
<pre class="example">     summary(glm(y~x+z, family=binomial))
 3801 
<pre class="example">     summary(glm(y~x+z, family=binomial))
3797 
     </pre>
 3802 
     </pre>
3798 
 
 3803 
 
3799 
<p>fits a logistic regression model, computes some summary information and
 3804 
<p>fits a logistic regression model, computes some summary information and
3800 
returns an object of class <code>"summary.glm"</code> (see <a href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>).
 3805 
returns an object of class <code>"summary.glm"</code> (see <a href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>).
3801 
 
 3806 
 
3802 
   <p>If you type <code>1+1</code> or <code>summary(glm(y~x+z, family=binomial))</code> at
 3807 
   <p>If you type <code>1+1</code> or <code>summary(glm(y~x+z, family=binomial))</code> at
3803 
the command line the returned value is automatically printed (unless it
 3808 
the command line the returned value is automatically printed (unless it
3804 
is <code>invisible()</code>), but in other circumstances, such as in a
 3809 
is <code>invisible()</code>), but in other circumstances, such as in a
3805 
<code>source()</code>d file or inside a function it isn't printed unless you
 3810 
<code>source()</code>d file or inside a function it isn't printed unless you
3806 
specifically print it.
 3811 
specifically print it.
3807 
 
 3812 
 
3808 
   <p>To print the value use
 3813 
   <p>To print the value use
3809 
 
 3814 
 
3810 
<pre class="example">     print(1+1)
 3815 
<pre class="example">     print(1+1)
3811 
     </pre>
 3816 
     </pre>
3812 
 
 3817 
 
3813 
<p>or
 3818 
<p>or
3814 
 
 3819 
 
3815 
<pre class="example">     print(summary(glm(y~x+z, family=binomial)))
 3820 
<pre class="example">     print(summary(glm(y~x+z, family=binomial)))
3816 
     </pre>
 3821 
     </pre>
3817 
 
 3822 
 
3818 
<p>instead, or use <code>source(</code><var>file</var><code>, echo=TRUE)</code>.
 3823 
<p>instead, or use <code>source(</code><var>file</var><code>, echo=TRUE)</code>.
3819 
 
 3824 
 
3820 
<div class="node">
 3825 
<div class="node">
3821 
<p><hr>
 3826 
<p><hr>
3822 
Node:&nbsp;<a name="Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
 3827 
Node:&nbsp;<a name="Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
3823 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
 3828 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
3824 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
 3829 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20is%20the%20output%20not%20printed%20when%20I%20source()%20a%20file%3f">Why is the output not printed when I source() a file?</a>,
3825 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3830 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3826 
<br>
 3831 
<br>
3827 
</div>
 3832 
</div>
3828 
 
 3833 
 
3829 
<h3 class="section">7.19 Why does outer() behave strangely with my function?</h3>
 3834 
<h3 class="section">7.19 Why does outer() behave strangely with my function?</h3>
3830 
 
 3835 
 
3831 
<p>As the help for <code>outer()</code> indicates, it does not work on arbitrary
 3836 
<p>As the help for <code>outer()</code> indicates, it does not work on arbitrary
3832 
functions the way the <code>apply()</code> family does.  It requires functions
 3837 
functions the way the <code>apply()</code> family does.  It requires functions
3833 
that are vectorized to work elementwise on arrays.  As you can see by
 3838 
that are vectorized to work elementwise on arrays.  As you can see by
3834 
looking at the code, <code>outer(x, y, FUN)</code> creates two large vectors
 3839 
looking at the code, <code>outer(x, y, FUN)</code> creates two large vectors
3835 
containing every possible combination of elements of <code>x</code> and
 3840 
containing every possible combination of elements of <code>x</code> and
3836 
<code>y</code> and then passes this to <code>FUN</code> all at once.  Your function
 3841 
<code>y</code> and then passes this to <code>FUN</code> all at once.  Your function
3837 
probably cannot handle two large vectors as parameters.
 3842 
probably cannot handle two large vectors as parameters.
3838 
 
 3843 
 
3839 
   <p>If you have a function that cannot handle two vectors but can handle two
 3844 
   <p>If you have a function that cannot handle two vectors but can handle two
3840 
scalars, then you can still use <code>outer()</code> but you will need to wrap
 3845 
scalars, then you can still use <code>outer()</code> but you will need to wrap
3841 
your function up first, to simulate vectorized behavior.  Suppose your
 3846 
your function up first, to simulate vectorized behavior.  Suppose your
3842 
function is
 3847 
function is
3843 
 
 3848 
 
3844 
<pre class="example">     foo &lt;- function(x, y, happy) {
 3849 
<pre class="example">     foo &lt;- function(x, y, happy) {
3845 
       stopifnot(length(x) == 1, length(y) == 1) # scalars only!
 3850 
       stopifnot(length(x) == 1, length(y) == 1) # scalars only!
3846 
       (x + y) * happy
 3851 
       (x + y) * happy
3847 
     }
 3852 
     }
3848 
     </pre>
 3853 
     </pre>
3849 
 
 3854 
 
3850 
<p>If you define the general function
 3855 
<p>If you define the general function
3851 
 
 3856 
 
3852 
<pre class="example">     wrapper &lt;- function(x, y, my.fun, ...) {
 3857 
<pre class="example">     wrapper &lt;- function(x, y, my.fun, ...) {
3853 
       sapply(seq(along = x), FUN = function(i) my.fun(x[i], y[i], ...))
 3858 
       sapply(seq(along = x), FUN = function(i) my.fun(x[i], y[i], ...))
3854 
     }
 3859 
     }
3855 
     </pre>
 3860 
     </pre>
3856 
 
 3861 
 
3857 
<p>then you can use <code>outer()</code> by writing, e.g.,
 3862 
<p>then you can use <code>outer()</code> by writing, e.g.,
3858 
 
 3863 
 
3859 
<pre class="example">     outer(1:4, 1:2, FUN = wrapper, my.fun = foo, happy = 10)
 3864 
<pre class="example">     outer(1:4, 1:2, FUN = wrapper, my.fun = foo, happy = 10)
3860 
     </pre>
 3865 
     </pre>
3861 
 
 3866 
 
3862 
<div class="node">
 3867 
<div class="node">
3863 
<p><hr>
 3868 
<p><hr>
3864 
Node:&nbsp;<a name="Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
 3869 
Node:&nbsp;<a name="Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
3865 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
 3870 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
3866 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
 3871 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20outer()%20behave%20strangely%20with%20my%20function%3f">Why does outer() behave strangely with my function?</a>,
3867 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3872 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3868 
<br>
 3873 
<br>
3869 
</div>
 3874 
</div>
3870 
 
 3875 
 
3871 
<h3 class="section">7.20 Why does the output from anova() depend on the order of factors in the model?</h3>
 3876 
<h3 class="section">7.20 Why does the output from anova() depend on the order of factors in the model?</h3>
3872 
 
 3877 
 
3873 
<p>In a model such as <code>~A+B+A:B</code>, R will report the difference in sums
 3878 
<p>In a model such as <code>~A+B+A:B</code>, R will report the difference in sums
3874 
of squares between the models <code>~1</code>, <code>~A</code>, <code>~A+B</code> and
 3879 
of squares between the models <code>~1</code>, <code>~A</code>, <code>~A+B</code> and
3875 
<code>~A+B+A:B</code>.  If the model were <code>~B+A+A:B</code>, R would report
 3880 
<code>~A+B+A:B</code>.  If the model were <code>~B+A+A:B</code>, R would report
3876 
differences between <code>~1</code>, <code>~B</code>, <code>~A+B</code>, and
 3881 
differences between <code>~1</code>, <code>~B</code>, <code>~A+B</code>, and
3877 
<code>~A+B+A:B</code> . In the first case the sum of squares for <code>A</code> is
 3882 
<code>~A+B+A:B</code> . In the first case the sum of squares for <code>A</code> is
3878 
comparing <code>~1</code> and <code>~A</code>, in the second case it is comparing
 3883 
comparing <code>~1</code> and <code>~A</code>, in the second case it is comparing
3879 
<code>~B</code> and <code>~B+A</code>.  In a non-orthogonal design (i.e., most
 3884 
<code>~B</code> and <code>~B+A</code>.  In a non-orthogonal design (i.e., most
3880 
unbalanced designs) these comparisons are (conceptually and numerically)
 3885 
unbalanced designs) these comparisons are (conceptually and numerically)
3881 
different.
 3886 
different.
3882 
 
 3887 
 
3883 
   <p>Some packages report instead the sums of squares based on comparing the
 3888 
   <p>Some packages report instead the sums of squares based on comparing the
3884 
full model to the models with each factor removed one at a time (the
 3889 
full model to the models with each factor removed one at a time (the
3885 
famous `Type III sums of squares' from SAS, for example).  These do not
 3890 
famous `Type III sums of squares' from SAS, for example).  These do not
3886 
depend on the order of factors in the model.  The question of which set
 3891 
depend on the order of factors in the model.  The question of which set
3887 
of sums of squares is the Right Thing provokes low-level holy wars on
 3892 
of sums of squares is the Right Thing provokes low-level holy wars on
3888 
R-help from time to time.
 3893 
R-help from time to time.
3889 
 
 3894 
 
3890 
   <p>There is no need to be agitated about the particular sums of squares
 3895 
   <p>There is no need to be agitated about the particular sums of squares
3891 
that R reports.  You can compute your favorite sums of squares quite
 3896 
that R reports.  You can compute your favorite sums of squares quite
3892 
easily.  Any two models can be compared with <code>anova(</code><var>model1</var><code>,
 3897 
easily.  Any two models can be compared with <code>anova(</code><var>model1</var><code>,
3893 
</code><var>model2</var><code>)</code>, and <code>drop1(</code><var>model1</var><code>)</code> will show the sums of
 3898 
</code><var>model2</var><code>)</code>, and <code>drop1(</code><var>model1</var><code>)</code> will show the sums of
3894 
squares resulting from dropping single terms.
 3899 
squares resulting from dropping single terms.
3895 
 
 3900 
 
3896 
<div class="node">
 3901 
<div class="node">
3897 
<p><hr>
 3902 
<p><hr>
3898 
Node:&nbsp;<a name="How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
 3903 
Node:&nbsp;<a name="How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
3899 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
 3904 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
3900 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
 3905 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20does%20the%20output%20from%20anova()%20depend%20on%20the%20order%20of%20factors%20in%20the%20model%3f">Why does the output from anova() depend on the order of factors in the model?</a>,
3901 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3906 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3902 
<br>
 3907 
<br>
3903 
</div>
 3908 
</div>
3904 
 
 3909 
 
3905 
<h3 class="section">7.21 How do I produce PNG graphics in batch mode?</h3>
 3910 
<h3 class="section">7.21 How do I produce PNG graphics in batch mode?</h3>
3906 
 
 3911 
 
3907 
<p>Under Unix, the <code>png()</code> device uses the X11 driver, which is a
 3912 
<p>Under Unix, the <code>png()</code> device uses the X11 driver, which is a
3908 
problem in batch mode or for remote operation.  If you have Ghostscript
 3913 
problem in batch mode or for remote operation.  If you have Ghostscript
3909 
you can use <code>bitmap()</code>, which produces a PostScript file then
 3914 
you can use <code>bitmap()</code>, which produces a PostScript file then
3910 
converts it to any bitmap format supported by ghostscript.  On some
 3915 
converts it to any bitmap format supported by ghostscript.  On some
3911 
installations this produces ugly output, on others it is perfectly
 3916 
installations this produces ugly output, on others it is perfectly
3912 
satisfactory.  In theory one could also use Xvfb, which provides an X
 3917 
satisfactory.  In theory one could also use Xvfb, which provides an X
3913 
server with no display.
 3918 
server with no display.
3914 
 
 3919 
 
3915 
<div class="node">
 3920 
<div class="node">
3916 
<p><hr>
 3921 
<p><hr>
3917 
Node:&nbsp;<a name="How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
 3922 
Node:&nbsp;<a name="How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
3918 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
 3923 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
3919 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
 3924 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20do%20I%20produce%20PNG%20graphics%20in%20batch%20mode%3f">How do I produce PNG graphics in batch mode?</a>,
3920 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3925 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3921 
<br>
 3926 
<br>
3922 
</div>
 3927 
</div>
3923 
 
 3928 
 
3924 
<h3 class="section">7.22 How can I get command line editing to work?</h3>
 3929 
<h3 class="section">7.22 How can I get command line editing to work?</h3>
3925 
 
 3930 
 
3926 
<p>The Unix command-line interface to R can only provide the inbuilt
 3931 
<p>The Unix command-line interface to R can only provide the inbuilt
3927 
command line editor which allows recall, editing and re-submission of
 3932 
command line editor which allows recall, editing and re-submission of
3928 
prior commands provided that the <small>GNU</small> readline library is
 3933 
prior commands provided that the <small>GNU</small> readline library is
3929 
available at the time R is configured for compilation.  Note that the
 3934 
available at the time R is configured for compilation.  Note that the
3930 
`development' version of readline including the appropriate headers is
 3935 
`development' version of readline including the appropriate headers is
3931 
needed: users of Linux binary distributions will need to install
 3936 
needed: users of Linux binary distributions will need to install
3932 
packages such as <code>libreadline-dev</code> (Debian) or
 3937 
packages such as <code>libreadline-dev</code> (Debian) or
3933 
<code>readline-devel</code> (Red Hat).
 3938 
<code>readline-devel</code> (Red Hat).
3934 
 
 3939 
 
3935 
<div class="node">
 3940 
<div class="node">
3936 
<p><hr>
 3941 
<p><hr>
3937 
Node:&nbsp;<a name="How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
 3942 
Node:&nbsp;<a name="How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
3938 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
 3943 
Next:&nbsp;<a rel="next" accesskey="n" href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
3939 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
 3944 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20get%20command%20line%20editing%20to%20work%3f">How can I get command line editing to work?</a>,
3940 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 3945 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3941 
<br>
 3946 
<br>
3942 
</div>
 3947 
</div>
3943 
 
 3948 
 
3944 
<h3 class="section">7.23 How can I turn a string into a variable?</h3>
 3949 
<h3 class="section">7.23 How can I turn a string into a variable?</h3>
3945 
 
 3950 
 
3946 
<p>If you have
 3951 
<p>If you have
3947 
 
 3952 
 
3948 
<pre class="example">     varname &lt;- c("a", "b", "d")
 3953 
<pre class="example">     varname &lt;- c("a", "b", "d")
3949 
     </pre>
 3954 
     </pre>
3950 
 
 3955 
 
3951 
<p>you can do
 3956 
<p>you can do
3952 
 
 3957 
 
3953 
<pre class="example">     get(varname[1]) + 2
 3958 
<pre class="example">     get(varname[1]) + 2
3954 
     </pre>
 3959 
     </pre>
3955 
 
 3960 
 
3956 
<p>for
 3961 
<p>for
3957 
 
 3962 
 
3958 
<pre class="example">     a + 2
 3963 
<pre class="example">     a + 2
3959 
     </pre>
 3964 
     </pre>
3960 
 
 3965 
 
3961 
<p>or
 3966 
<p>or
3962 
 
 3967 
 
3963 
<pre class="example">     assign(varname[1], 2 + 2)
 3968 
<pre class="example">     assign(varname[1], 2 + 2)
3964 
     </pre>
 3969 
     </pre>
3965 
 
 3970 
 
3966 
<p>for
 3971 
<p>for
3967 
 
 3972 
 
3968 
<pre class="example">     a &lt;- 2 + 2
 3973 
<pre class="example">     a &lt;- 2 + 2
3969 
     </pre>
 3974 
     </pre>
3970 
 
 3975 
 
3971 
<p>or
 3976 
<p>or
3972 
 
 3977 
 
3973 
<pre class="example">     eval(substitute(lm(y ~ x + variable),
 3978 
<pre class="example">     eval(substitute(lm(y ~ x + variable),
3974 
                     list(variable = as.name(varname[1]))
 3979 
                     list(variable = as.name(varname[1]))
3975 
     </pre>
 3980 
     </pre>
3976 
 
 3981 
 
3977 
<p>for
 3982 
<p>for
3978 
 
 3983 
 
3979 
<pre class="example">     lm(y ~ x + a)
 3984 
<pre class="example">     lm(y ~ x + a)
3980 
     </pre>
 3985 
     </pre>
3981 
 
 3986 
 
3982 
   <p>At least in the first two cases it is often easier to just use a list,
 3987 
   <p>At least in the first two cases it is often easier to just use a list,
3983 
and then you can easily index it by name
 3988 
and then you can easily index it by name
3984 
 
 3989 
 
3985 
<pre class="example">     vars &lt;- list(a = 1:10, b = rnorm(100), d = LETTERS)
 3990 
<pre class="example">     vars &lt;- list(a = 1:10, b = rnorm(100), d = LETTERS)
3986 
     vars[["a"]]
 3991 
     vars[["a"]]
3987 
     </pre>
 3992 
     </pre>
3988 
 
 3993 
 
3989 
<p>without any of this messing about.
 3994 
<p>without any of this messing about.
3990 
 
 3995 
 
3991 
<div class="node">
 3996 
<div class="node">
3992 
<p><hr>
 3997 
<p><hr>
3993 
Node:&nbsp;<a name="Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
 3998 
Node:&nbsp;<a name="Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
3994 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>,
 3999 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>,
3995 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
 4000 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20turn%20a%20string%20into%20a%20variable%3f">How can I turn a string into a variable?</a>,
3996 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 4001 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
3997 
<br>
 4002 
<br>
3998 
</div>
 4003 
</div>
3999 
 
 4004 
 
4000 
<h3 class="section">7.24 Why do lattice/trellis graphics not work?</h3>
 4005 
<h3 class="section">7.24 Why do lattice/trellis graphics not work?</h3>
4001 
 
 4006 
 
4002 
<p>The most likely reason is that you forgot to tell R to display the
 4007 
<p>The most likely reason is that you forgot to tell R to display the
4003 
graph.  Lattice functions such as <code>xyplot()</code> create a graph object,
 4008 
graph.  Lattice functions such as <code>xyplot()</code> create a graph object,
4004 
but do not display it (the same is true of Trellis graphics in
 4009 
but do not display it (the same is true of Trellis graphics in
4005 
<small>S-PLUS</small>).  The <code>print()</code> method for the graph object produces the
 4010 
<small>S-PLUS</small>).  The <code>print()</code> method for the graph object produces the
4006 
actual display.  When you use these functions interactively at the
 4011 
actual display.  When you use these functions interactively at the
4007 
command line, the result is automatically printed, but in
 4012 
command line, the result is automatically printed, but in
4008 
<code>source()</code> or inside your own functions you will need an explicit
 4013 
<code>source()</code> or inside your own functions you will need an explicit
4009 
<code>print()</code> statement.
 4014 
<code>print()</code> statement.
4010 
 
 4015 
 
4011 
<div class="node">
 4016 
<div class="node">
4012 
<p><hr>
 4017 
<p><hr>
4013 
Node:&nbsp;<a name="How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>,
 4018 
Node:&nbsp;<a name="How%20can%20I%20sort%20the%20rows%20of%20a%20data%20frame%3f">How can I sort the rows of a data frame?</a>,
4014 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
 4019 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#Why%20do%20lattice%2ftrellis%20graphics%20not%20work%3f">Why do lattice/trellis graphics not work?</a>,
4015 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
 4020 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Miscellanea">R Miscellanea</a>
4016 
<br>
 4021 
<br>
4017 
</div>
 4022 
</div>
4018 
 
 4023 
 
4019 
<h3 class="section">7.25 How can I sort the rows of a data frame?</h3>
 4024 
<h3 class="section">7.25 How can I sort the rows of a data frame?</h3>
4020 
 
 4025 
 
4021 
<p>To sort the rows within a data frame, with respect to the values in one
 4026 
<p>To sort the rows within a data frame, with respect to the values in one
4022 
or more of the columns, simply use <code>order()</code>.
 4027 
or more of the columns, simply use <code>order()</code>.
4023 
 
 4028 
 
4024 
<div class="node">
 4029 
<div class="node">
4025 
<p><hr>
 4030 
<p><hr>
4026 
Node:&nbsp;<a name="R%20Programming">R Programming</a>,
 4031 
Node:&nbsp;<a name="R%20Programming">R Programming</a>,
4027 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Bugs">R Bugs</a>,
 4032 
Next:&nbsp;<a rel="next" accesskey="n" href="#R%20Bugs">R Bugs</a>,
4028 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Miscellanea">R Miscellanea</a>,
 4033 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Miscellanea">R Miscellanea</a>,
4029 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 4034 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
4030 
<br>
 4035 
<br>
4031 
</div>
 4036 
</div>
4032 
 
 4037 
 
4033 
<h2 class="chapter">8 R Programming</h2>
 4038 
<h2 class="chapter">8 R Programming</h2>
4034 
 
 4039 
 
4035 
<ul class="menu">
 4040 
<ul class="menu">
4036 
<li><a accesskey="1" href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>:
4041 
<li><a accesskey="1" href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>:
4037 
<li><a accesskey="2" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>:
4042 
<li><a accesskey="2" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>:
4038 
<li><a accesskey="3" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>:
4043 
<li><a accesskey="3" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>:
4039 
<li><a accesskey="4" href="#How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>:
4044 
<li><a accesskey="4" href="#How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>:
4040 
</ul>
 4045 
</ul>
4041 
 
 4046 
 
4042 
<div class="node">
 4047 
<div class="node">
4043 
<p><hr>
 4048 
<p><hr>
4044 
Node:&nbsp;<a name="How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
 4049 
Node:&nbsp;<a name="How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
4045 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
 4050 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
4046 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Programming">R Programming</a>,
 4051 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Programming">R Programming</a>,
4047 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
 4052 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
4048 
<br>
 4053 
<br>
4049 
</div>
 4054 
</div>
4050 
 
 4055 
 
4051 
<h3 class="section">8.1 How should I write summary methods?</h3>
 4056 
<h3 class="section">8.1 How should I write summary methods?</h3>
4052 
 
 4057 
 
4053 
<p>Suppose you want to provide a summary method for class <code>"foo"</code>.
4058 
<p>Suppose you want to provide a summary method for class <code>"foo"</code>.
4054 
Then <code>summary.foo()</code> should not print anything, but return an
 4059 
Then <code>summary.foo()</code> should not print anything, but return an
4055 
object of class <code>"summary.foo"</code>, <em>and</em> you should write a
 4060 
object of class <code>"summary.foo"</code>, <em>and</em> you should write a
4056 
method <code>print.summary.foo()</code> which nicely prints the summary
 4061 
method <code>print.summary.foo()</code> which nicely prints the summary
4057 
information and invisibly returns its object.  This approach is
 4062 
information and invisibly returns its object.  This approach is
4058 
preferred over having <code>summary.foo()</code> print summary information and
 4063 
preferred over having <code>summary.foo()</code> print summary information and
4059 
return something useful, as sometimes you need to grab something
 4064 
return something useful, as sometimes you need to grab something
4060 
computed by <code>summary()</code> inside a function or similar.  In such
 4065 
computed by <code>summary()</code> inside a function or similar.  In such
4061 
cases you don't want anything printed.
 4066 
cases you don't want anything printed.
4062 
 
 4067 
 
4063 
<div class="node">
 4068 
<div class="node">
4064 
<p><hr>
 4069 
<p><hr>
4065 
Node:&nbsp;<a name="How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
 4070 
Node:&nbsp;<a name="How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
4066 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
 4071 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
4067 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
 4072 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
4068 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
 4073 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
4069 
<br>
 4074 
<br>
4070 
</div>
 4075 
</div>
4071 
 
 4076 
 
4072 
<h3 class="section">8.2 How can I debug dynamically loaded code?</h3>
 4077 
<h3 class="section">8.2 How can I debug dynamically loaded code?</h3>
4073 
 
 4078 
 
4074 
<p>Roughly speaking, you need to start R inside the debugger, load the
 4079 
<p>Roughly speaking, you need to start R inside the debugger, load the
4075 
code, send an interrupt, and then set the required breakpoints.
 4080 
code, send an interrupt, and then set the required breakpoints.
4076 
 
 4081 
 
4077 
   <p>See section "Finding entry points in dynamically loaded code" in
 4082 
   <p>See section "Finding entry points in dynamically loaded code" in
4078 
<cite>Writing R Extensions</cite>.
4083 
<cite>Writing R Extensions</cite>.
4079 
This manual is included in the R distribution, see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>.
 4084 
This manual is included in the R distribution, see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>.
4080 
 
 4085 
 
4081 
<div class="node">
 4086 
<div class="node">
4082 
<p><hr>
 4087 
<p><hr>
4083 
Node:&nbsp;<a name="How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
 4088 
Node:&nbsp;<a name="How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
4084 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>,
 4089 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>,
4085 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
 4090 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
4086 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
 4091 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
4087 
<br>
 4092 
<br>
4088 
</div>
 4093 
</div>
4089 
 
 4094 
 
4090 
<h3 class="section">8.3 How can I inspect R objects when debugging?</h3>
 4095 
<h3 class="section">8.3 How can I inspect R objects when debugging?</h3>
4091 
 
 4096 
 
4092 
<p>The most convenient way is to call <code>R_PV</code> from the symbolic
 4097 
<p>The most convenient way is to call <code>R_PV</code> from the symbolic
4093 
debugger.
 4098 
debugger.
4094 
 
 4099 
 
4095 
   <p>See section "Inspecting R objects when debugging" in <cite>Writing R
 4100 
   <p>See section "Inspecting R objects when debugging" in <cite>Writing R
4096 
Extensions</cite>.
 4101 
Extensions</cite>.
4097 
 
 4102 
 
4098 
<div class="node">
 4103 
<div class="node">
4099 
<p><hr>
 4104 
<p><hr>
4100 
Node:&nbsp;<a name="How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>,
 4105 
Node:&nbsp;<a name="How%20can%20I%20change%20compilation%20flags%3f">How can I change compilation flags?</a>,
4101 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
 4106 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
4102 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
 4107 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Programming">R Programming</a>
4103 
<br>
 4108 
<br>
4104 
</div>
 4109 
</div>
4105 
 
 4110 
 
4106 
<h3 class="section">8.4 How can I change compilation flags?</h3>
 4111 
<h3 class="section">8.4 How can I change compilation flags?</h3>
4107 
 
 4112 
 
4108 
<p>Suppose you have C code file for dynloading into R, but you want to use
 4113 
<p>Suppose you have C code file for dynloading into R, but you want to use
4109 
<code>R CMD SHLIB</code> with compilation flags other than the default ones
 4114 
<code>R CMD SHLIB</code> with compilation flags other than the default ones
4110 
(which were determined when R was built).  You could change the file
 4115 
(which were determined when R was built).  You could change the file
4111 
<code>R_HOME/etc/Makeconf</code> to reflect your preferences.  If you
 4116 
<code>R_HOME/etc/Makeconf</code> to reflect your preferences.  If you
4112 
are a Bourne shell user, you can also pass the desired flags to Make
 4117 
are a Bourne shell user, you can also pass the desired flags to Make
4113 
(which is used for controlling compilation) via the Make variable
 4118 
(which is used for controlling compilation) via the Make variable
4114 
<code>MAKEFLAGS</code>, as in
 4119 
<code>MAKEFLAGS</code>, as in
4115 
 
 4120 
 
4116 
<pre class="example">     MAKEFLAGS="CFLAGS=-O3" R CMD SHLIB *.c
 4121 
<pre class="example">     MAKEFLAGS="CFLAGS=-O3" R CMD SHLIB *.c
4117 
     </pre>
 4122 
     </pre>
4118 
 
 4123 
 
4119 
<div class="node">
 4124 
<div class="node">
4120 
<p><hr>
 4125 
<p><hr>
4121 
Node:&nbsp;<a name="R%20Bugs">R Bugs</a>,
 4126 
Node:&nbsp;<a name="R%20Bugs">R Bugs</a>,
4122 
Next:&nbsp;<a rel="next" accesskey="n" href="#Acknowledgments">Acknowledgments</a>,
 4127 
Next:&nbsp;<a rel="next" accesskey="n" href="#Acknowledgments">Acknowledgments</a>,
4123 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Programming">R Programming</a>,
 4128 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Programming">R Programming</a>,
4124 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 4129 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
4125 
<br>
 4130 
<br>
4126 
</div>
 4131 
</div>
4127 
 
 4132 
 
4128 
<h2 class="chapter">9 R Bugs</h2>
 4133 
<h2 class="chapter">9 R Bugs</h2>
4129 
 
 4134 
 
4130 
<ul class="menu">
 4135 
<ul class="menu">
4131 
<li><a accesskey="1" href="#What%20is%20a%20bug%3f">What is a bug?</a>:
4136 
<li><a accesskey="1" href="#What%20is%20a%20bug%3f">What is a bug?</a>:
4132 
<li><a accesskey="2" href="#How%20to%20report%20a%20bug">How to report a bug</a>:
4137 
<li><a accesskey="2" href="#How%20to%20report%20a%20bug">How to report a bug</a>:
4133 
</ul>
 4138 
</ul>
4134 
 
 4139 
 
4135 
<div class="node">
 4140 
<div class="node">
4136 
<p><hr>
 4141 
<p><hr>
4137 
Node:&nbsp;<a name="What%20is%20a%20bug%3f">What is a bug?</a>,
 4142 
Node:&nbsp;<a name="What%20is%20a%20bug%3f">What is a bug?</a>,
4138 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20to%20report%20a%20bug">How to report a bug</a>,
 4143 
Next:&nbsp;<a rel="next" accesskey="n" href="#How%20to%20report%20a%20bug">How to report a bug</a>,
4139 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Bugs">R Bugs</a>,
 4144 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Bugs">R Bugs</a>,
4140 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Bugs">R Bugs</a>
 4145 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Bugs">R Bugs</a>
4141 
<br>
 4146 
<br>
4142 
</div>
 4147 
</div>
4143 
 
 4148 
 
4144 
<h3 class="section">9.1 What is a bug?</h3>
 4149 
<h3 class="section">9.1 What is a bug?</h3>
4145 
 
 4150 
 
4146 
<p>If R executes an illegal instruction, or dies with an operating system
 4151 
<p>If R executes an illegal instruction, or dies with an operating system
4147 
error message that indicates a problem in the program (as opposed to
 4152 
error message that indicates a problem in the program (as opposed to
4148 
something like "disk full"), then it is certainly a bug.  If you call
 4153 
something like "disk full"), then it is certainly a bug.  If you call
4149 
<code>.C()</code>, <code>.Fortran()</code>, <code>.External()</code> or <code>.Call()</code> (or
 4154 
<code>.C()</code>, <code>.Fortran()</code>, <code>.External()</code> or <code>.Call()</code> (or
4150 
<code>.Internal()</code>) yourself (or in a function you wrote), you can
 4155 
<code>.Internal()</code>) yourself (or in a function you wrote), you can
4151 
always crash R by using wrong argument types (modes).  This is not a
 4156 
always crash R by using wrong argument types (modes).  This is not a
4152 
bug.
 4157 
bug.
4153 
 
 4158 
 
4154 
   <p>Taking forever to complete a command can be a bug, but you must make
 4159 
   <p>Taking forever to complete a command can be a bug, but you must make
4155 
certain that it was really R's fault.  Some commands simply take a long
 4160 
certain that it was really R's fault.  Some commands simply take a long
4156 
time.  If the input was such that you <em>know</em> it should have been
 4161 
time.  If the input was such that you <em>know</em> it should have been
4157 
processed quickly, report a bug.  If you don't know whether the command
 4162 
processed quickly, report a bug.  If you don't know whether the command
4158 
should take a long time, find out by looking in the manual or by asking
 4163 
should take a long time, find out by looking in the manual or by asking
4159 
for assistance.
 4164 
for assistance.
4160 
 
 4165 
 
4161 
   <p>If a command you are familiar with causes an R error message in a case
 4166 
   <p>If a command you are familiar with causes an R error message in a case
4162 
where its usual definition ought to be reasonable, it is probably a bug.
4167 
where its usual definition ought to be reasonable, it is probably a bug.
4163 
If a command does the wrong thing, that is a bug.  But be sure you know
 4168 
If a command does the wrong thing, that is a bug.  But be sure you know
4164 
for certain what it ought to have done.  If you aren't familiar with the
 4169 
for certain what it ought to have done.  If you aren't familiar with the
4165 
command, or don't know for certain how the command is supposed to work,
 4170 
command, or don't know for certain how the command is supposed to work,
4166 
then it might actually be working right.  Rather than jumping to
 4171 
then it might actually be working right.  Rather than jumping to
4167 
conclusions, show the problem to someone who knows for certain.
 4172 
conclusions, show the problem to someone who knows for certain.
4168 
 
 4173 
 
4169 
   <p>Finally, a command's intended definition may not be best for statistical
 4174 
   <p>Finally, a command's intended definition may not be best for statistical
4170 
analysis.  This is a very important sort of problem, but it is also a
 4175 
analysis.  This is a very important sort of problem, but it is also a
4171 
matter of judgment.  Also, it is easy to come to such a conclusion out
 4176 
matter of judgment.  Also, it is easy to come to such a conclusion out
4172 
of ignorance of some of the existing features.  It is probably best not
 4177 
of ignorance of some of the existing features.  It is probably best not
4173 
to complain about such a problem until you have checked the
 4178 
to complain about such a problem until you have checked the
4174 
documentation in the usual ways, feel confident that you understand it,
 4179 
documentation in the usual ways, feel confident that you understand it,
4175 
and know for certain that what you want is not available.  If you are
 4180 
and know for certain that what you want is not available.  If you are
4176 
not sure what the command is supposed to do after a careful reading of
 4181 
not sure what the command is supposed to do after a careful reading of
4177 
the manual this indicates a bug in the manual.  The manual's job is to
 4182 
the manual this indicates a bug in the manual.  The manual's job is to
4178 
make everything clear.  It is just as important to report documentation
 4183 
make everything clear.  It is just as important to report documentation
4179 
bugs as program bugs.  However, we know that the introductory
 4184 
bugs as program bugs.  However, we know that the introductory
4180 
documentation is seriously inadequate, so you don't need to report this.
 4185 
documentation is seriously inadequate, so you don't need to report this.
4181 
 
 4186 
 
4182 
   <p>If the online argument list of a function disagrees with the manual, one
 4187 
   <p>If the online argument list of a function disagrees with the manual, one
4183 
of them must be wrong, so report the bug.
 4188 
of them must be wrong, so report the bug.
4184 
 
 4189 
 
4185 
<div class="node">
 4190 
<div class="node">
4186 
<p><hr>
 4191 
<p><hr>
4187 
Node:&nbsp;<a name="How%20to%20report%20a%20bug">How to report a bug</a>,
 4192 
Node:&nbsp;<a name="How%20to%20report%20a%20bug">How to report a bug</a>,
4188 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20a%20bug%3f">What is a bug?</a>,
 4193 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#What%20is%20a%20bug%3f">What is a bug?</a>,
4189 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Bugs">R Bugs</a>
 4194 
Up:&nbsp;<a rel="up" accesskey="u" href="#R%20Bugs">R Bugs</a>
4190 
<br>
 4195 
<br>
4191 
</div>
 4196 
</div>
4192 
 
 4197 
 
4193 
<h3 class="section">9.2 How to report a bug</h3>
 4198 
<h3 class="section">9.2 How to report a bug</h3>
4194 
 
 4199 
 
4195 
<p>When you decide that there is a bug, it is important to report it and to
 4200 
<p>When you decide that there is a bug, it is important to report it and to
4196 
report it in a way which is useful.  What is most useful is an exact
 4201 
report it in a way which is useful.  What is most useful is an exact
4197 
description of what commands you type, starting with the shell command
 4202 
description of what commands you type, starting with the shell command
4198 
to run R, until the problem happens.  Always include the version of R,
 4203 
to run R, until the problem happens.  Always include the version of R,
4199 
machine, and operating system that you are using; type <kbd>version</kbd> in
 4204 
machine, and operating system that you are using; type <kbd>version</kbd> in
4200 
R to print this.
 4205 
R to print this.
4201 
 
 4206 
 
4202 
   <p>The most important principle in reporting a bug is to report
 4207 
   <p>The most important principle in reporting a bug is to report
4203 
<em>facts</em>, not hypotheses or categorizations.  It is always easier to
 4208 
<em>facts</em>, not hypotheses or categorizations.  It is always easier to
4204 
report the facts, but people seem to prefer to strain to posit
 4209 
report the facts, but people seem to prefer to strain to posit
4205 
explanations and report them instead.  If the explanations are based on
 4210 
explanations and report them instead.  If the explanations are based on
4206 
guesses about how R is implemented, they will be useless; others will
 4211 
guesses about how R is implemented, they will be useless; others will
4207 
have to try to figure out what the facts must have been to lead to such
 4212 
have to try to figure out what the facts must have been to lead to such
4208 
speculations.  Sometimes this is impossible.  But in any case, it is
 4213 
speculations.  Sometimes this is impossible.  But in any case, it is
4209 
unnecessary work for the ones trying to fix the problem.
 4214 
unnecessary work for the ones trying to fix the problem.
4210 
 
 4215 
 
4211 
   <p>For example, suppose that on a data set which you know to be quite large
 4216 
   <p>For example, suppose that on a data set which you know to be quite large
4212 
the command
 4217 
the command
4213 
 
 4218 
 
4214 
<pre class="example">     R&gt; data.frame(x, y, z, monday, tuesday)
 4219 
<pre class="example">     R&gt; data.frame(x, y, z, monday, tuesday)
4215 
     </pre>
 4220 
     </pre>
4216 
 
 4221 
 
4217 
<p>never returns.  Do not report that <code>data.frame()</code> fails for large
 4222 
<p>never returns.  Do not report that <code>data.frame()</code> fails for large
4218 
data sets.  Perhaps it fails when a variable name is a day of the week.
4223 
data sets.  Perhaps it fails when a variable name is a day of the week.
4219 
If this is so then when others got your report they would try out the
 4224 
If this is so then when others got your report they would try out the
4220 
<code>data.frame()</code> command on a large data set, probably with no day of
 4225 
<code>data.frame()</code> command on a large data set, probably with no day of
4221 
the week variable name, and not see any problem.  There is no way in the
 4226 
the week variable name, and not see any problem.  There is no way in the
4222 
world that others could guess that they should try a day of the week
 4227 
world that others could guess that they should try a day of the week
4223 
variable name.
 4228 
variable name.
4224 
 
 4229 
 
4225 
   <p>Or perhaps the command fails because the last command you used was a
 4230 
   <p>Or perhaps the command fails because the last command you used was a
4226 
method for <code>"["()</code> that had a bug causing R's internal data
 4231 
method for <code>"["()</code> that had a bug causing R's internal data
4227 
structures to be corrupted and making the <code>data.frame()</code> command
 4232 
structures to be corrupted and making the <code>data.frame()</code> command
4228 
fail from then on.  This is why others need to know what other commands
 4233 
fail from then on.  This is why others need to know what other commands
4229 
you have typed (or read from your startup file).
 4234 
you have typed (or read from your startup file).
4230 
 
 4235 
 
4231 
   <p>It is very useful to try and find simple examples that produce
 4236 
   <p>It is very useful to try and find simple examples that produce
4232 
apparently the same bug, and somewhat useful to find simple examples
 4237 
apparently the same bug, and somewhat useful to find simple examples
4233 
that might be expected to produce the bug but actually do not.  If you
 4238 
that might be expected to produce the bug but actually do not.  If you
4234 
want to debug the problem and find exactly what caused it, that is
 4239 
want to debug the problem and find exactly what caused it, that is
4235 
wonderful.  You should still report the facts as well as any
 4240 
wonderful.  You should still report the facts as well as any
4236 
explanations or solutions.  Please include an example that reproduces
 4241 
explanations or solutions.  Please include an example that reproduces
4237 
the problem, preferably the simplest one you have found.
 4242 
the problem, preferably the simplest one you have found.
4238 
 
 4243 
 
4239 
   <p>Invoking R with the <code>--vanilla</code> option may help in isolating a
 4244 
   <p>Invoking R with the <code>--vanilla</code> option may help in isolating a
4240 
bug.  This ensures that the site profile and saved data files are not
 4245 
bug.  This ensures that the site profile and saved data files are not
4241 
read.
 4246 
read.
4242 
 
 4247 
 
4243 
   <p>On Unix systems a bug report can be generated using the function
 4248 
   <p>On Unix systems a bug report can be generated using the function
4244 
<code>bug.report()</code>.  This automatically includes the version
 4249 
<code>bug.report()</code>.  This automatically includes the version
4245 
information and sends the bug to the correct address.  Alternatively the
 4250 
information and sends the bug to the correct address.  Alternatively the
4246 
bug report can be emailed to <a href="mailto:R-bugs@R-project.org">R-bugs@R-project.org</a> or submitted
 4251 
bug report can be emailed to <a href="mailto:R-bugs@R-project.org">R-bugs@R-project.org</a> or submitted
4247 
to the Web page at <a href="http://bugs.R-project.org/">http://bugs.R-project.org/</a>.
 4252 
to the Web page at <a href="http://bugs.R-project.org/">http://bugs.R-project.org/</a>.
4248 
 
 4253 
 
4249 
   <p>Bug reports on contributed packages should perhaps be sent to the
 4254 
   <p>Bug reports on contributed packages should perhaps be sent to the
4250 
package maintainer rather than to R-bugs.
 4255 
package maintainer rather than to R-bugs.
4251 
 
 4256 
 
4252 
<div class="node">
 4257 
<div class="node">
4253 
<p><hr>
 4258 
<p><hr>
4254 
Node:&nbsp;<a name="Acknowledgments">Acknowledgments</a>,
 4259 
Node:&nbsp;<a name="Acknowledgments">Acknowledgments</a>,
4255 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Bugs">R Bugs</a>,
 4260 
Previous:&nbsp;<a rel="previous" accesskey="p" href="#R%20Bugs">R Bugs</a>,
4256 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
 4261 
Up:&nbsp;<a rel="up" accesskey="u" href="#Top">Top</a>
4257 
<br>
 4262 
<br>
4258 
</div>
 4263 
</div>
4259 
 
 4264 
 
4260 
<h2 class="chapter">10 Acknowledgments</h2>
 4265 
<h2 class="chapter">10 Acknowledgments</h2>
4261 
 
 4266 
 
4262 
<p>Of course, many many thanks to Robert and Ross for the R system, and to
 4267 
<p>Of course, many many thanks to Robert and Ross for the R system, and to
4263 
the package writers and porters for adding to it.
 4268 
the package writers and porters for adding to it.
4264 
 
 4269 
 
4265 
   <p>Special thanks go to Doug Bates, Peter Dalgaard, Paul Gilbert, Stefano
 4270 
   <p>Special thanks go to Doug Bates, Peter Dalgaard, Paul Gilbert, Stefano
4266 
Iacus, Fritz Leisch, Jim Lindsey, Thomas Lumley, Martin Maechler, Brian
 4271 
Iacus, Fritz Leisch, Jim Lindsey, Thomas Lumley, Martin Maechler, Brian
4267 
D. Ripley, Anthony Rossini, and Andreas Weingessel for their comments
 4272 
D. Ripley, Anthony Rossini, and Andreas Weingessel for their comments
4268 
which helped me improve this <small>FAQ</small>.
 4273 
which helped me improve this <small>FAQ</small>.
4269 
 
 4274 
 
4270 
   <p>More to some soon <small class="dots">...</small>
 4275 
   <p>More to some soon <small class="dots">...</small>
4271 
 
 4276 
 
4272 
   </body></html>
 4277 
   </body></html>
4273 
 
 4278