I will try to make this more precise. In the lme() function, the "correlation" argument allows the user to specify a within-group correlation structure, i.e. the structure of the Lambda matrix using the mixed model notation in Pineiro and Bates. What I want to do is specify a distinct structure for the Psi matrix (same notation), that is, a correlation structure for the random effects.
If lme() doesn't allow for this, is there any other function that I could use? MUHC-Research wrote: > > Dear R users, > > I would be interested in using the lme() function to fit a linear mixed > model to a longitudinal dataset. I know this function allows for the > specification of a within-group covariance structure. However, does it > allow for the explicit specification of a between-group covariance > structure? > > Being able to specify both separately would be very important in the > context of my project since, as might be expected, they have different > implications/interpretations. > > For instance, the mixed procedure in SAS allows users to specify the two > structures separately by adding a value for the type argument after the > RANDOM statement and the REPEATED statement. > > My question is thus if we can do the same with lme(). > > I thank you most sincerely for your help. > -- View this message in context: http://www.nabble.com/lme-between-group-and-within-group-covariance-tp22834748p22868945.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
