Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | ZhangVic <stataquestion@hotmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: biprobit with partial observability+IV |
Date | Tue, 6 Aug 2013 09:43:54 +0000 |
Hi statalisters, I need to run a bivariate probit model with partial observability since I only observe the product of Y1 and Y2 (two binary dependent variable). However, the problem I have now is I have a continuous endogenous regressor in each equation. However, there is no existing command like ivprobit to conduct such biprobit with both IV and partial observability. I am thinking to use a two-step procedure manually. Running an OLS (Z=a0+a1X+ controls) to get the predicted of Z (my instrument) first. Then, replace my x in the biprobit with Z_hat. But, I still need to correct the standard errors of my regressors. I find an exactly the same question posted in statalist: http://www.stata.com/statalist/archive/2009-09/msg00814.html However, no one answers her. Anyone knows how to correct the standard errors in this case? Thank you in advance!! Best, Vic * * 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/