Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: getting u0 from -xtlogit, re-


From   Jeph Herrin <jherrin@flyingbuttress.net>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: getting u0 from -xtlogit, re-
Date   Wed, 28 Feb 2007 17:55:44 -0500

There is an error in the second paragraph, where I should
have written:

 > -predict- after *-xtlogit-* allows a linear prediction which
 > assumes u0 = 0, and a probability prediction which assumes
 > u0=0.

Jeph Herrin wrote:
-xtlogit, re- estimates a random intercept logistic
regression model, and returns estimates of the variance
of the random (group) error term u0 in two different ways:
ln(sigma_u^2) and sigma_u.

However, unlike with -xtmixed-, there seems to be no
direct way to generate predictions that incorporate u0.
-predict- after -xtmixed- allows a linear prediction which
assumes u0 = 0, and a probability prediction which assumes
u0=0. What I want is a predicted value which equals the
linear prediction plus the random effect.

Given that -predict- doesn't do this, does someone know a
trick for getting u0 for each group? Or a theoretical reason
that I shouldn't be able to? I think I could roll my own,
but it would be easier to detour through a software package
which will give me these directly.

thanks,
Jeph

*
*   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/


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index