[R-sig-ME] Empirical variance estimator of SE's, lmer

Afshartous, David DAfshartous at med.miami.edu
Mon Jun 22 16:28:05 CEST 2009


I searched the archives and didn't find anything on empirical variance
estimates of SEs of fixed effects in lmer, i.e., replacing Var(Y_i) with
(y_i - X_i \hat{beta}) in the standard formula for Var(\hat{beta}).  Does
code exist for this?  It should only be a few lines but was just wondering
if it already existed.


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