Dear statalist,
I am trying to estimate a system of 3
equations(rather 4 equations as will become clear). I
will write the model down with slight abuse of
notations(not differentiating clearly between latent
and observed variables)
The first equation(or first 2 equations) is a
standard Heckman model:
Y1=X1b1+u1 and a selection equation Y0=X0b0+u0 where
Y1 is observed only if Y0>Y*.
The second equation can be one of two types:
1)A censored data model(tobit).
Y2=a2Y1+X2b2+u2 where we only observe Y2>0
2) A heterogenous count data model where in the above
equation Y2 only takes 0 or positive integer values or
explicitly
Y2=exp(b12Y1+X2b22+u2)
And finally a linear equation:
Y3=a3log(Y2)+X3b3+u3
Is there a command or program in stata which allows me
to estimate such a system? For example can we estimate
a heckman model simultaneouly with a tobit and a
linear equation? Or more generally is there a
program/command in stata which can estimate a
simultaneous system of equations when one of the
equation in the system is characterised by the heckman
model?Thank you.
Thoihen
__________________________________
Do you Yahoo!?
Take Yahoo! Mail with you! Get it on your mobile phone.
http://mobile.yahoo.com/maildemo
*
* 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/