--- Kristian Jakobsen wrote
> I have bivariate probit model where I model the transition in and out
> of poverty between two periods. I am however only interested in the
> probability of being poor in period 2(either staying poor or entering
> poverty). That means that the two transitions that I am interested in
> are the probability of being poor in period 2 conditional of being
> poor in period 1 together with the probability of being poor in
> period 2 conditional on being non-poor in period 2. More formally,
> I have:
>
> y1: poverty status in period 1
> y2: poverty status in period 2
> hh: vector of household characteristics
>
> But how would you model that in Stata with the conditionalities. I
> have put in:
>
> biprobit y1 y2 hh
>
> But as far as I can see this will not estimate the probabilities that
> I want.
The trivial solution is to use -probit y2 hh-, if all you care
about is y2, than model y2. Alternatively, you can recover the
probabilities you are after after -biprobit- using -predict- with
the -pmarg2- option, see -help biprobit postestimation##predict-.
*-------- begin example -------------------
webuse school
biprobit private vote logptax loginc years
predict prvote , pmarg2
mfx, predict(pmarg2)
*---------- end example -------------------
( For more on how to use examples I sent to statalist see:
http://www.maartenbuis.nl/stata/exampleFAQ.html )
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/
