GLMM: Fit Generalized Linear Mixed Models via PQL

Description

Fit a GLMM model with multivariate normal random effects, using Penalized Quasi-Likelihood.

Usage

GLMM(formula, family, data, random, control, niter, method, verbose, ...)

Arguments

formula

a two-sided linear model formula giving fixed-effects part of the model.

family

a GLM family, see glm.

random

A formula or list of formulae describing the random effects.

data

an optional data frame used as the first place to find variables in the formulae.

control

an optional argument to be passed to lme.

niter

maximum number of PQL iterations. Default is 20.

method

character: Estimation method to be used. Possible values are "PQL", the default, or "Laplace". "PQL" provides penalized quasi-likelihood estimates. "Laplace" provides PQL estimation foll

verbose

logical: print out record of iterations? Default is FALSE.

...

Optional further arguments such as subset and na.action.

Value

A object of class "glmm": see reStruct-class.

References

Schall, R. (1991) Estimation in generalized linear models with random effects. Biometrika 78, 719--727.

Breslow, N. E. and Clayton, D. G. (1993) Approximate inference in generalized linear mixed models. Journal of the American Statistical Association 88, 9--25.

Wolfinger, R. and O'Connell, M. (1993) Generalized linear mixed models: a pseudo-likelihood approach. Journal of Statistical Computation and Simulation 48, 233--243.

Examples

Run this code

library(lme4)
data(guImmun)
fm1 = GLMM(immun ~ kid2p + mom25p + ord + ethn +
                  momEd + husEd + momWork + rural + pcInd81,
          data = guImmun, family = binomial,
          random = ~1|comm/mom)
summary(fm1)
fm2 = GLMM(immun ~ kid2p + mom25p + ord + ethn +
                  momEd + husEd + momWork + rural + pcInd81,
          data = guImmun, family = binomial,
          random = ~1|comm, method = 'Laplace')
summary(fm2)