st: Simultaneous bivariate probit model

[Sakib Mahmud <smahmud@uwyo.edu>]

Dear all, 

I have a simultaneous bivariate probit model. Y1 and Y2 are related to each other. Among the exogenous variables, X1 is common to both equations. But the other regressors are different. The basic structure of the model is,  

Y1 = a0 + a1Y2 + a2X1 + a3X2 + a4X3 + u1
Y2 = b0 + b1Y1 + b2X1+b3 X4 +b4X5 + u2

I found out that usually we use -biprobit- command in STATA for seemingly unrelated bivariate probit model. But since I am dealing with simultaneous bivariate probit model should I use the -biprobit- as well? Do I need to use -reg3- ? If yes, then what is the correct command specification to deal with simultaneous bivariate probit model? 

Currently, I am thinking about using the following command, 
.biprobit (Y1=Y2 X1 X2 X3) (Y2=Y1 X1 X4 X5), robust

But I think this is not the correct way to approach my problem. 

I will be glad if you can guide me to apply the correct specification. 

Thank you, 
Sakib 
PhD student
University of Wyoming


*
*   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/