ranef.glmm.admb: Extract Random Effects

In bbolker/glmmadmb: Generalized Linear Mixed Models using 'AD Model Builder'

Description

Extract random effects from fitted glmmadmb models, and (optionally) their standard deviations.

Usage

## S3 method for class 'glmmadmb'
ranef(object, sd=FALSE, scale=TRUE, ...)

Arguments

object	a fitted glmmadmb model.
sd	whether the standard deviations of the random effects should be returned, rather than point estimates.
scale	(logical) scale random effects by the relevant random-effects standard deviations (i.e., the resulting effects should have variance approximately equal to the estimated RE variance if scale=TRUE, or approximately equal to 1 if scale=FALSE)
...	further arguments (currently ignored).

Details

The point estimates of the random effects are ‘empirical Bayes’ estimates in the sense that the hyperparameters (fixed effects, variance components, etc.) are fixed at their MLE's when the joint likelihood is maximized with respect to the random effects. However, uncertainty about the hyperparameters is taken into account when the standard deviations of the random effects are calculated. The standard deviations of the random effects are quadratic approximations, so should be taken with a grain of salt for small sample sizes.

Value

A matrix of estimated random effects (or standard deviations).

See Also

glmmadmb.

glmmADMB-package.

bbolker/glmmadmb documentation built on May 11, 2019, 9:29 p.m.