mclogit.fit mclogit 0.8.5.1

Internal functions used for model fit.

Description

These functions are exported and documented for use by other packages. They are not intended for end users.

Usage

mclogit.fit(y, s, w, X,
            dispersion=FALSE,
            start = NULL, offset = NULL,
            control = mclogit.control())

mmclogit.fitPQLMQL(y, s, w, X, Z, groups, start,
                   offset = NULL, method=c("PQL","MQL"),
                   estimator = c("ML","REML"),
                   control = mmclogit.control())

Arguments

y

a response vector. Should be binary.

s

a vector identifying individuals or covariate strata

w

a vector with observation weights.

X

a model matrix; required.

dispersion

a logical value or a character string; whether and how a dispersion parameter should be estimated. For details see dispersion.

Z

the random effects design matrix.

groups

a list of grouping factors.

start

an optional numerical vector of starting values for the coefficients.

offset

an optional model offset. Currently only supported for models without random effects.

method

a character string, either “PQL” or “MQL”, specifies the type of the quasilikelihood approximation.

estimator

a character string; either “ML” or “REML”, specifies which estimator is to be used/approximated.

control

a list of parameters for the fitting process. See mclogit.control

Value

A list with components describing the fitted model.