On 9/24/06, Nishant Dass <[email protected]> wrote:
Dear list-members,
I would like to estimate a logit model with sample
selection but since there's only -heckprob- available in
Stata, I was wondering whether it would be incorrect to do
the following instead:
1. Estimate the selection equation using -logit-.
2. Calculate the inverse Mills' ratio (IMR).
3. Use the IMR from above as a regressor in a second-stage
logit model.
I wonder if this approach is correct, i.e. if there is a theoretical
justification for it. I have never seen one (but this may reflect my
ignorance ...;-)) Even if the approach is correct, you will have to
correct the standard errors in your second step regression.
A much better approach would be to choose a probit model. In Stata you
could then use the -heckprob- command.
Hope this helps,
Jean
*
* 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/
