To get empirical Bayes predictions of the random slopes
(and intercepts), use the 'u' option:
gllapred reffs, u
will produce new variables reffsm1 reffsm2 etc.
containing the predicted values of random effect
1 (e.g. intercepts), 2 (slopes), etc.
You also get posterior standard deviations in
reffs1, reffs2, etc.
I hope this helps,
Sophia
----- Original Message -----
From: "Michael McQuestion" <[email protected]>
To: <[email protected]>
Cc: "Feng Zhao" <[email protected]>
Sent: Wednesday, September 03, 2003 3:21 PM
Subject: st: Retrieving specific slopes using gllapred
> Dear Statalisters,
>
> How can one retrieve specific slope estimates following estimation of a
gllamm model? To clarify, after fitting a two-level variance components
model, I need to retrieve a second-level beta, Bj, and use it as a predictor
in a second model.
>
> Thanks in advance.
>
> Michael J. McQuestion
> Assistant Professor
> Johns Hopkins Bloomberg School of Public Health
> Department of Population and Family Health Sciences
> 615 Wolfe Street, Rm. E-4142
> Baltimore, MD 21205
>
> phone (410) 502-6037
> fax (410) 955-2303
>
> Only those who will risk going too far can possibly find out how far they
can go.
> T.S. Eliot
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
