For R, I understand that the package lme4 and the function glmer roughly corresponds to glimmix in SAS. What is the default covariance structure when fit and can it be changed? If so how?
$\begingroup$
$\endgroup$
1
-
4$\begingroup$ Check out this answer: stats.stackexchange.com/questions/5344/… $\endgroup$– fabiansCommented Feb 24, 2011 at 12:43
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
I do not know about SAS, but variance in glmer is controlled by family argument. If you want to change correlation structure then I suspect you will have to use nlme from nlme package.
