Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Danny Dan <danny2011dan@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Simultaneous Equation Model |
Date | Tue, 31 May 2011 08:40:39 -0500 |
Hello, I am estimating a simultaneous equation probit model with two equations as following: Y1=a0+a1*Y2+a2*X1+e1 Y2=b0+b1*Y1+b2*X2+e2 where both Y1 and Y2 are binary. Y1 and Y2 are related and endogenous. X1 and X2 are sets of exogenous variables and e1 and e2 error terms. X1 and X2 may have some common variables but not all for the purpose of identification. There is a method described in Madalla (1983) (model 6, Ch#8,Page# 247) in order to estimate this type of models. But I do not know how to compute the variance-covariance matrix using the same. The formula for calculating variance and covariance matrix uses PDF (Probability density function) and CDF (Cumulative density function) but it is not mentioned how to calculate those after the probit estimation. My questions are: (i) Is there any code in Stata which can help estimating this model (like CDSIMEQ). (ii) Can I use 'biprobit' in order to estimate this model? (iii) Is there any other way to estimate this model? Please help. Thank you in advance. Dan * * 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/