Dear all,
I'm using a first order markov model, on panel data, to model poverty transitions and control for the initial conditions problem. In stata I use a biprobit model and I would like to control for unobserved heterogeneity. I have thought of using random effects or creating a variable that would capture the time invariant unobservable individual specific effects. However it appears neither of the options are possible in stata. Do you know any way I could control for that?
Dimitris Souftas
_________________________________________________________________
More than messages–check out the rest of the Windows Live™.
http://www.microsoft.com/windows/windowslive/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
