st: mvprobit

[Nyasha Tirivayi <ntirivayi@gmail.com>]

Hello

I am now trying to estimate a model on the effects of a food aid
program on labour market dynamics of HIV affected households. Receipt
of food aid is also an endogenous variable, while I am also trying to
deal with the initial conditions problem. I have panel data from
pre-food aid and 6 months after food aid program began. I am also
suspecting that this could involve 3 recursive simultanaeous equations
as follows:


Employment at t  =  f (foodaid, age, gender, household income, education)

Food aid =  f (age, gender, education, household income, HIV
prevalence in area, household dependency ratio, employment at t-1)

Employment at t-1  =  f (age, gender, household income, education)


I am thinking of using mvprobit or triprobit for transition analysis
controlling for initial conditions and the endogeneity of food aid. I
am especially interested in doing separate sub-group analysis e.g.
looking at the transitions for those who were employed pre-program and
those who were unemployed.

Would you kindly advise on how best to write my code for people who
made transition from unemployment to employment? How do I write the
code to specifically estimate the transition to employment conditional
on previous emplorment state? Do I include employment at t-1 in the
current employment equation as a lagged explanatory variable?


Kindly advise

Nyasha Tirivayi
Maastricht University

*
*   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/