Hi all statalist readers:
I am estimating the following two-equation model:
Y1 = a1X1 + B1Y2 + e1
Y2*= a2X2 + e2
where Y2=1 if Y2*>0
Y2=0 otherwise
Y1 is censored at zero and Y2 is binary (the realised value of the latent
variable
Y2*). Since Y2 is assumed to be endogenous I first obtained a prediction for
Y2 from a probit regression of Y2 on several regressors thought to influence
Y2. This predicted value was then used in a tobit second stage estimation of
Y1. Just to check, I also run a tobit for Y1 using simply a dummy for Y2,
and the coeffients obtained were different in magnitud. I wonder if the fact
of obtaining different results is in some way indicative that the first
approach is correct, i.e. using a prediction for Y2 instead of a dummy. Of
course I tested the endogeneity of Y2, but I am not sure if the estimation
approach is really correcting for the presumed endogeneity of Y2.
Thanks in advance for any help.
Ricardo Henr�quez
Chile
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
