summary.glmfm {fit.models} | R Documentation |
Comparison Summaries for Generalized Linear Models
Description
Compute a summary of each model in a glmfm
object.
Usage
## S3 method for class 'glmfm'
summary(object, correlation = FALSE, ...)
Arguments
object |
a glmfm object. |
correlation |
a logical value. If |
... |
additional arguments required by the generic
|
Value
a list with class summary.glmfm whose elements are summaries of each
model in object
.
Examples
# From ?glm:
# A Gamma example, from McCullagh & Nelder (1989, pp. 300-2)
clotting <- data.frame(
u = c(5,10,15,20,30,40,60,80,100),
lot1 = c(118,58,42,35,27,25,21,19,18),
lot2 = c(69,35,26,21,18,16,13,12,12))
lot1 <- glm(lot1 ~ log(u), data = clotting, family = Gamma)
lot2 <- glm(lot2 ~ log(u), data = clotting, family = Gamma)
fm <- fit.models(lot1, lot2)
summary(fm)
[Package fit.models version 0.64 Index]