The
summary.glm method is applied
to each
glm component
of
object to produce summary information
on the individual fits,
which is organized into a list of summary statistics.
The returned object is suitable for printing
with the
print.summary.glmList method.
USAGE:
summary(object, pool)
REQUIRED ARGUMENTS:
object
an object inheriting from class
glmList,
representing a list of
glm fitted objects.
pool
an optional logical value indicating whether a pooled
estimate of the residual deviance should be used.
Default is
attr(object, "pool").
VALUE:
a list with summary statistics obtained
by applying
summary.glm to the elements
of
object,
inheriting from class
summary.glmList.
The components of
value are:
call
a list containing an image of the
glmList
call that produced
object.
coefficients
a three dimensional array with summary information
on the
glm coefficients.
The first dimension corresponds to the names
of the
object components,
the second dimension is given by
"Value",
"Std. Error"
,
"t value"
,
and
"Pr(>|t|)",
corresponding, respectively, to the coefficient estimates
and their associated standard errors, t-values, and p-values.
The third dimension is given by the coefficients names.
deviance
if
pool=TRUE,
the sum of the residual deviances
from all
glm fits;
else, an array with the residual deviances
of the individual
glm fits.
df.residual
the total number of degrees of freedom for residuals,
corresponding to the sum of residuals df of all
glm
components.
dispersion
if
pool=TRUE,
the pooled estimate of dispersion parameter;
else, an array with the estimated dispersions
of the individual
glm fits.
null.deviance
if
pool=TRUE,
the sum of the null deviances from all
glm fits;
else, an array with the null deviances
of the individual
glm fits.
pool
the value of the
pool argument to the function.
terms
the terms object used in fitting the individual
glm
components.
SEE ALSO:
,
.
EXAMPLES:
fm1 <- glmList(resp ~ trt | clinic, Clinic, family=binomial)
summary(fm1)