Statalist The Stata Listserver


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

st: GLLAMM - probabilities


From   "Ellen Van de Poel" <vandepoel@few.eur.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: GLLAMM - probabilities
Date   Fri, 30 Mar 2007 15:04:57 +1000

Dear all,

I am estimating a model of child mortality (1-0) using household and
community random effects. So I have cross-sectional data with a binary
outcome and three levels: child, household community.

I am using the following command:

gllamm u1yrd $xvars , i(hhcode PSU) pweight(wt) f(binom) l(logit) adapt
trace

Thereafter I predict the posterior means of the random effects and the
probabilities:

gllapred re, u 

gllapred p0, mu 

Now, my problem is that I don't understand why the p0's that I estimate are
not the same as what I get from

gllapred test , mu us(rem)

I would like to make predictions by changing the value of the random
effects. But I'm not sure how to do this since it seems that I don't get the
predicted probability with the latter two equtions.

Could anyone please help me in this matter?

Many thanks in advance for any suggestion,

Ellen Van de Poel



*
*   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