logLik.mlmm: Log-Likelihood From Multiple Linear Mixed Models

In bozenne/repeated: Repeated Measurement Models for Discrete Times

Log-Likelihood From Multiple Linear Mixed Models

Description

Extract or compute the log-likelihood from each group-specific Linear Mixed Model.

Usage

## S3 method for class 'mlmm'
logLik(object, newdata = NULL, p = NULL, indiv = FALSE, ...)

Arguments

object	a mlmm object.
newdata	[data.frame] dataset relative to which the log-likelihood should be computed. Only relevant if differs from the dataset used to fit the model.
p	[list of numeric vector] values for the model parameters at which to evaluate the log-likelihood. Only relevant if differs from the fitted values.
indiv	[logical] Should the contribution of each cluster to the log-likelihood be output? Otherwise output the sum of all clusters of the derivatives.
...	Not used. For compatibility with the generic method.

Details

indiv: only relevant when using maximum likelihood. Must be FALSE when using restricted maximum likelihood.

Value

A numeric vector when indiv=FALSE (log-likelihood per model) or a matrix of numeric values when when indiv=TRUE with one line for each cluster and one column for each model.

bozenne/repeated documentation built on July 16, 2025, 11:16 p.m.