Dear Statalist,
It seems to me that there is no equivalent of the "heckman" command for
a sample selection problem in which the model for the selection equation
is Tobit, rather than Probit. Is that correct?
I would like to estimate such a model, and I know that I can do it
manually using Stata (i.e. by (1) using Stata's MLE capabilities or (2)
by estimating the selection and primary equations in 2 steps, with the
appropriate corrections for sample selection).
But if there is an equivalent of the "heckman" command that will do the
2-step procedure automatically (particularly the correction of the
standard errors of the primary equation estimates), then naturally that
would be preferable.
I've searched the Stata documentation, and I don't see a command or an
option for "heckman" that would do the trick.
Thanks!
Steve
*
* 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/
