Summary Method for Generalized Linear Models

DESCRIPTION:

Returns a summary of the generalized linear models stored in a fit.models object with virtual class glmfm.

USAGE:

summary.glmfm(object, correlation=T, ...)

REQUIRED ARGUMENTS:

object
a fit.models object with virtual class glmfm that contains the fitted models.

OPTIONAL ARGUMENTS:

correlation
if TRUE, the correlation matrix for the coefficients is included in the summary.

VALUE:

a summary.lmfm object with the following components:
mod.names
the names of the models in object.
calls
a list containing the call of each model in object.
restats
a 2-dimensional array with one row (containing the minimum, the maximum, and the quartiles of the residuals) for each model in object.
coefs
a 3-dimensional array. The third index corresponds to the models in object. Each 2-dimensional slice is a matrix with one row for each term in the model and columns for the coefficient estimate, its standard error, the t statistic, and the p-value.
df
a vector containing the residual degrees of freedom for each model in object.
devs
a vector containing the residual deviance for each model in object.
correlations
a 3-dimensional array. For each level of the 3rd index (one for each model in object), the 2-dimensional slice is the estimated correlation matrix of the coefficients. This will not be present if correlation is FALSE.

SEE ALSO:

, , , ,

EXAMPLES:

breslow.mle <- glm(sumY ~ Age10 + Base4 * Trt, family=poisson,
     data=breslow.dat)
breslow.rob <- glmRob(sumY ~ Age10 + Base4 * Trt, family=poisson,
     data=breslow.dat)
breslow.fm <- fit.models(breslow.rob, breslow.mle)
summary(breslow.fm)