| Line 21... |
Line 21... |
| 21 |
both be factors.}
|
21 |
both be factors.}
|
| 22 |
\item{y}{a numeric vector; ignored if \code{x} is a matrix. If
|
22 |
\item{y}{a numeric vector; ignored if \code{x} is a matrix. If
|
| 23 |
\code{x} is a factor, \code{y} should be a factor of the same length.}
|
23 |
\code{x} is a factor, \code{y} should be a factor of the same length.}
|
| 24 |
\item{correct}{a logical indicating whether to apply continuity
|
24 |
\item{correct}{a logical indicating whether to apply continuity
|
| 25 |
correction when computing the test statistic for 2 by 2 tables: one
|
25 |
correction when computing the test statistic for 2 by 2 tables: one
|
| 26 |
half is subtracted from all \eqn{|O - E|} differences; however, the
|
26 |
half is subtracted from all \eqn{|O - E|} differences; however, the
|
| 27 |
correction will not be bigger than the differences themselves. No correction
|
27 |
correction will not be bigger than the differences themselves. No correction
|
| 28 |
is done if \code{simulate.p.value = TRUE}.}
|
28 |
is done if \code{simulate.p.value = TRUE}.}
|
| 29 |
\item{p}{a vector of probabilities of the same length of \code{x}.
|
29 |
\item{p}{a vector of probabilities of the same length of \code{x}.
|
| 30 |
An error is given if any entry of \code{p} is negative.}
|
30 |
An error is given if any entry of \code{p} is negative.}
|
| 31 |
\item{rescale.p}{a logical scalar; if TRUE then \code{p} is rescaled
|
31 |
\item{rescale.p}{a logical scalar; if TRUE then \code{p} is rescaled
|
| Line 88... |
Line 88... |
| 88 |
\item{observed}{the observed counts.}
|
88 |
\item{observed}{the observed counts.}
|
| 89 |
\item{expected}{the expected counts under the null hypothesis.}
|
89 |
\item{expected}{the expected counts under the null hypothesis.}
|
| 90 |
\item{residuals}{the Pearson residuals,
|
90 |
\item{residuals}{the Pearson residuals,
|
| 91 |
\code{(observed - expected) / sqrt(expected)}.}
|
91 |
\code{(observed - expected) / sqrt(expected)}.}
|
| 92 |
\item{stdres}{standardized residuals,
|
92 |
\item{stdres}{standardized residuals,
|
| 93 |
\code{(observed - expected) / sqrt(V)}, where \code{V} is the residual cell variance (Agresti, 2007,
|
93 |
\code{(observed - expected) / sqrt(V)}, where \code{V} is the residual cell variance (Agresti, 2007,
|
| 94 |
section 2.4.5 for the case where \code{x} is a matrix, \code{n * p * (1 - p)} otherwise).}
|
94 |
section 2.4.5 for the case where \code{x} is a matrix, \code{n * p * (1 - p)} otherwise).}
|
| 95 |
}
|
95 |
}
|
| 96 |
\seealso{
|
96 |
\seealso{
|
| 97 |
For goodness-of-fit testing, notably of continuous distributions,
|
97 |
For goodness-of-fit testing, notably of continuous distributions,
|
| 98 |
\code{\link{ks.test}}.
|
98 |
\code{\link{ks.test}}.
|
| Line 122... |
Line 122... |
| 122 |
## From Agresti(2007) p.39
|
122 |
## From Agresti(2007) p.39
|
| 123 |
M <- as.table(rbind(c(762, 327, 468), c(484, 239, 477)))
|
123 |
M <- as.table(rbind(c(762, 327, 468), c(484, 239, 477)))
|
| 124 |
dimnames(M) <- list(gender = c("M","F"),
|
124 |
dimnames(M) <- list(gender = c("M","F"),
|
| 125 |
party = c("Democrat","Independent", "Republican"))
|
125 |
party = c("Democrat","Independent", "Republican"))
|
| 126 |
(Xsq <- chisq.test(M)) # Prints test summary
|
126 |
(Xsq <- chisq.test(M)) # Prints test summary
|
| 127 |
Xsq$observed # observed counts (same as M)
|
127 |
Xsq$observed # observed counts (same as M)
|
| 128 |
Xsq$expected # expected counts under the null
|
128 |
Xsq$expected # expected counts under the null
|
| 129 |
Xsq$residuals # Pearson residuals
|
129 |
Xsq$residuals # Pearson residuals
|
| 130 |
Xsq$stdres # standardized residuals
|
130 |
Xsq$stdres # standardized residuals
|
| 131 |
|
131 |
|
| 132 |
|
132 |
|
The R Project SVN
R