Re: st: twostep option heckman

[Fernando Rios Avila <f.rios.a@gmail.com>]

Hi Chiara,
Keep in mind that the MLE estimates assume the errors from the probit
and ols are simultaneously normal. Whereas the twostep one relaxes
this assumption. As far as I know, mle is more efficient, but twostep
is consistent.
Hope this help.
HTH

On Wed, Nov 27, 2013 at 2:40 PM, Chiara Mussida <cmussida@gmail.com> wrote:
> dear all,
> if the heckman model for selection does not converge (not concave) the
> option twostep allows getting consistent mle estimates. Are these
> estimates reliable even if, as said, the model without the option
> twostep does not converge?
> Thanks for any kind of clarification.
> Best
> chiara
>
> --
> Chiara Mussida
> Assistant Professor
> Department of Economics and Social Sciences,
> Università Cattolica del Sacro Cuore,
> Piacenza (Italy)
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/