st: cmp command with bivariate probit model and hierarchical dataset

[Fatma M Romeh <fromeh1@student.gsu.edu>]

Dear statalist,

I have a small question about your -cmp- command. I'm trying to fit hierarchical/random effects  bivariate probit model to model school and work decisions jointly. There are three levels (individuals,  households, and regions). I have independent variables at  each of these levels and the set of these independent variables are similar for school and work equations. I read the -cmp- manual and tried a lot to fit the model, but I'm really having difficulty with that. Stata keeps ruining forever. I thought I might write it wrong. simply, this is what I'm typing:


cmp (work = $all_Xs || hh_id: || region_id:) (school= $all_Xs || hh_id: || region_id), ind($cmp_probit $cmp_probit) nolr


I'm new to the -cmp- command and would really appreciate any help on how to correctly write the syntax for my mode.


Thank you in advance,

Fatma,

    
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/