Class "glmm"

Description

A generalized linear mixed-effects model.

Objects from the Class

Objects can be created by calls of the form new("glmm", ...) but more commonly they are created by calls to GLMM.

Slots

family:

Object of class "family", specifying the error distribution and the link function used in the model.

origy:

Object of class "numeric", the original response.

n:

Object of class "numeric", if the family is binomial, the number of observations for each data point.

prior.weights:

Object of class "numeric", weights used when creating the model.frame.

init.weights:

Object of class "numeric", weights from a fixed effects generalized linear model for the data.

init.y:

Object of class "numeric", the linear predictor from a fixed effects generalized linear model for the data.

method:

Object of class "character", the method used to fit the generalized linear mixed model.

reStruct:

Object of class "reStruct", from class "lme", the random-effects structure for the model.

frame:

Object of class "data.frame", from class "lme", the model.frame used to fit the model.

na.action:

Object of class "ANY", from class "lme", the na.action argument used when creating the model.frame.

fitted:

Object of class "numeric", from class "lme", the fitted values in the linear predictor scale.

call:

Object of class "call", from class "lme", the function call used to create the object.

Extends

Class "lme", directly.

Methods

LMEoptimize<-

signature(x = "glmm", value = "list"): optimize the PQL approximation to the log-likelihood.

VarCorr

signature(x = "glmm"): Extract variance and correlation components.

fixef<-

signature(object = "glmm", value = "numeric"): assign the fixed effects (used for method = "Laplace").

getResponse

signature(object = "glmm"): extract the response.

logLik

signature(object = "glmm"): extract the (approximate) log-likelihood.

show

signature(object = "glmm"): show the object.

summary

signature(object = "glmm"): summarize the object.

Author(s)

Saikat DebRoy saikat@stat.wisc.edu and Douglas Bates bates@stat.wisc.edu

Examples

library(lme4)
data(guImmun)
# This returns an object of class glmm
fm = GLMM(immun ~ kid2p + mom25p + ord + ethn +
                  momEd + husEd + momWork + rural + pcInd81,
          data = guImmun, family = binomial,
          random = ~1|comm/mom)
fm