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/
