If it’s the variance of the random
effects you are after, try
VarCorr(fit.object)
stella
From:
s-news-owner@lists.biostat.wustl.edu
[mailto:s-news-owner@lists.biostat.wustl.edu] On
Behalf Of James Stapleton
Sent: Wednesday, May 14, 2008 1:54
PM
To: s-news@lists.biostat.wustl.edu
Subject: [S] lme
I am using S-Plus 8.0 (Windows), and the function “lme”
to fit a model with fixed and random components. The function works well. I
call the resulting object a.lme.
Then I define s.lme = summary(a.lme). Printing that gives me
what I need. However, I would like to simulate this many times, generating a
new y vector each time, and for each such iteration store the resulting
parameter estimates. I can do this for the estimates of the parameters (the betas) in the fixed part. However, neither of
the lists a.lme and s.lme contain the estimates of the standard deviations (or
variance) estimates, even though they are printed when I enter
s.lme or print(s.lme) ( which are the same).
How can I “grab” these estimates so that I can
store them?
Jim Stapleton