Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: heckprob and rho


From   Antoine Terracol <Antoine.Terracol@univ-paris1.fr>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: heckprob and rho
Date   Sat, 07 Sep 2002 18:39:04 +0200

Hello,
Since -heckprob- assumes normality of the residuals, you should use -probit- instead of -logit-.
Of course the results will not be exactly the same as in -heckprob- since rho will be constrained to equal zero.

hope this helps,
Antoine

M. Nizam Khan wrote:


Hi statausers,

I am trying to estimate "ever visit to health care provider" model with sample selection. My probit equation is visit=a+xb+u1 and selection equation is reported_sickness=c+zb+u2. In both cases, dependent variables are binary (1,0). Variables used for identification are: smoker, ADL, use of hygenic latrine.
These variables have very low correlation with "VISIT".
I used heckprob to estimate my model. What I would like to know: When rho is not significant, should I run a simple logit model and interpret my results based on that? I was expecting that when rho is not significant, coefficents and SE (of visit equation) in the simple logit procedure should not be different from heckprob procedure. In my case, I found it different. Any suggestion on this would be appreciated.

Thnaks.

Nizam *
* 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/



*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index