Dear all,
I want to estimate 2 equations using bivariate probit model . The problem is that I'm dealing with Panel data and while searching, I find that (biprobit) command can not be used with panel structure. Any idea about the code on Stata, which takes into account the panel structure using bivariate probit model ?
Thanks you in advance
It depends if you are using fixed effects or random effects. If it is random effects model you can try to optimize the likelihood using a Monte Carlo EM algorithm. You can code this algorithm on Matlab or R Studio, I am not sure about Stata.
Hey Doc,
As Imran says, I think you can use the "xtprobit" command with the usual data declaration in you analysis with "xtset". This way, Stata calls for all panel aided commands with the prefix "xt". Monka has a good suggestion as well especially if your two equations are system equations. This can be computationally challenging but I know you have used R and Matlab before. Either of these are sufficient to have the work done. If you need to work in Stata, you then will have to write a program under Stata's Mata.
Hi Dina Abdelsalam , you can use the following command for bivariate probit model for your panel model.
biprobit (DV1= IV1 IV2 IV3..IVn) (DV2= IV1 IV2 IV3), robust
In case you want to use bivariate probit model with partial observability, then use partial at the end of first command.
biprobit (DV1= IV1 IV2 IV3..IVn) (DV2= IV1 IV2 IV3), partial
I hope this answer give some help.
Dear Abdelsalam,
you can also refer David Roodman cmp stata command is also a very interesting and robust one!
- Badges
- Science topic
- Similar topics
- Filing