Re: st: Different selection variables for different subpopulations in the sample

[Dimitrije Tišma <dimitrijetisma@gmail.com>]

Thanks a lot Maarten, it sounds like a great idea!

Dimitrije




2011/4/17 Maarten buis <maartenbuis@yahoo.co.uk>:
> --- On Sun, 17/4/11, Dimitrije Tišma wrote:
>> I was wondering if there exists some Stata command which
>> would allow me to select different selection variables when
>> working with Heckman selection models. For instance, let's
>> say I am working with the textbook example of wages. I may
>> assume that number of kids is the variable that determines
>> selection of women into employment, while some other
>> variable would be selection variable for men. Or,
>> alternatively, I could assume that there is selection bias
>> only for women, while no selection is at work for the other
>> part of the sample.
>
> I would just add interaction terms between a female indicator
> variable and relevant other variables in your selection
> equation, i.e. I would not expect the number of children to
> have the same effect on the probability of working for men and
> women.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl
> --------------------------
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/