[R-sig-ME] Standard errors for random effects

Douglas Bates bates at stat.wisc.edu
Thu Apr 1 17:03:33 CEST 2010


On Thu, Apr 1, 2010 at 8:41 AM, Tahira Jamil <tahirajamil at yahoo.com> wrote:

> I am wondering how can we extract standard errors of random effects. Is it possible or not in lme4.

Technically the random effects don't have standard errors because they
are not parameters in the model.  If you use

ranef(mylmermodel, postVar = TRUE)

there will be an attribute attached to the random effects that is part
of the conditional variance-covariance matrix of the random effects.
It is used, for example, to calculate the prediction intervals on the
random effects shown in the dotplot or qqmath plots for them.  Check

library(lme4)
example(ranef)

or even

?ranef




More information about the R-sig-mixed-models mailing list