| Line 116... |
Line 116... |
| 116 |
extract from the fitted model object.}
|
116 |
extract from the fitted model object.}
|
| 117 |
|
117 |
|
| 118 |
\item{\dots}{
|
118 |
\item{\dots}{
|
| 119 |
For \code{glm}: arguments to be used to form the default
|
119 |
For \code{glm}: arguments to be used to form the default
|
| 120 |
\code{control} argument if it is not supplied directly.
|
120 |
\code{control} argument if it is not supplied directly.
|
| 121 |
|
121 |
|
| 122 |
For \code{weights}: further arguments passed to or from other methods.
|
122 |
For \code{weights}: further arguments passed to or from other methods.
|
| 123 |
}
|
123 |
}
|
| 124 |
}
|
124 |
}
|
| 125 |
\details{
|
125 |
\details{
|
| 126 |
A typical predictor has the form \code{response ~ terms} where
|
126 |
A typical predictor has the form \code{response ~ terms} where
|
| Line 190... |
Line 190... |
| 190 |
\value{
|
190 |
\value{
|
| 191 |
\code{glm} returns an object of class inheriting from \code{"glm"}
|
191 |
\code{glm} returns an object of class inheriting from \code{"glm"}
|
| 192 |
which inherits from the class \code{"lm"}. See later in this section.
|
192 |
which inherits from the class \code{"lm"}. See later in this section.
|
| 193 |
If a non-standard \code{method} is used, the object will also inherit
|
193 |
If a non-standard \code{method} is used, the object will also inherit
|
| 194 |
from the class (if any) returned by that function.
|
194 |
from the class (if any) returned by that function.
|
| 195 |
|
195 |
|
| 196 |
The function \code{\link{summary}} (i.e., \code{\link{summary.glm}}) can
|
196 |
The function \code{\link{summary}} (i.e., \code{\link{summary.glm}}) can
|
| 197 |
be used to obtain or print a summary of the results and the function
|
197 |
be used to obtain or print a summary of the results and the function
|
| 198 |
\code{\link{anova}} (i.e., \code{\link{anova.glm}})
|
198 |
\code{\link{anova}} (i.e., \code{\link{anova.glm}})
|
| 199 |
to produce an analysis of variance table.
|
199 |
to produce an analysis of variance table.
|
| 200 |
|
200 |
|
| Line 288... |
Line 288... |
| 288 |
\code{\link{effects}}, \code{\link{fitted.values}},
|
288 |
\code{\link{effects}}, \code{\link{fitted.values}},
|
| 289 |
and \code{\link{residuals}}.
|
289 |
and \code{\link{residuals}}.
|
| 290 |
|
290 |
|
| 291 |
\code{\link{lm}} for non-generalized \emph{linear} models (which SAS
|
291 |
\code{\link{lm}} for non-generalized \emph{linear} models (which SAS
|
| 292 |
calls GLMs, for \sQuote{general} linear models).
|
292 |
calls GLMs, for \sQuote{general} linear models).
|
| 293 |
|
293 |
|
| 294 |
\code{\link{loglin}} and \code{\link[MASS]{loglm}} (package
|
294 |
\code{\link{loglin}} and \code{\link[MASS]{loglm}} (package
|
| 295 |
\CRANpkg{MASS}) for fitting log-linear models (which binomial and
|
295 |
\CRANpkg{MASS}) for fitting log-linear models (which binomial and
|
| 296 |
Poisson GLMs are) to contingency tables.
|
296 |
Poisson GLMs are) to contingency tables.
|
| 297 |
|
297 |
|
| 298 |
\code{bigglm} in package \CRANpkg{biglm} for an alternative
|
298 |
\code{bigglm} in package \CRANpkg{biglm} for an alternative
|
| 299 |
way to fit GLMs to large datasets (especially those with many cases).
|
299 |
way to fit GLMs to large datasets (especially those with many cases).
|
| 300 |
|
300 |
|
| 301 |
\code{\link{esoph}}, \code{\link{infert}} and
|
301 |
\code{\link{esoph}}, \code{\link{infert}} and
|
| 302 |
\code{\link{predict.glm}} have examples of fitting binomial glms.
|
302 |
\code{\link{predict.glm}} have examples of fitting binomial glms.
|
| 303 |
}
|
303 |
}
|
| 304 |
\author{
|
304 |
\author{
|
| 305 |
The original \R implementation of \code{glm} was written by Simon
|
305 |
The original \R implementation of \code{glm} was written by Simon
|
The R Project SVN
R