[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: endogenous variable with missing
Dear Statalist members
I have a couple of questions.
I have the following system of simultaneous equations:
Q= b1*Zq + b2*W + e i=1....N
[Q observed over whole sample]
W= a1*Zw + a2*Q + u i=1...n < N
[W observed only for sub-sample i=1...n<N and missing
for the rest of observations ]
I want to estimate eqn Q for the whole sample. If I
use standard IV the command gives me estimate only for
the subsample for which W is observed. I have been
told to estimate a reduced form of equation W, then
predict W for the whole sample and insert it in the
second stage in the eqn for Q.
Could anyone confirm this?
Second, how should I proceed if D is actually censored
and I want estimated it by tobit?
thank you in advance for any suggestion
L'email della prossima generazione? Puoi averla con la nuova Yahoo! Mail:
* For searches and help try: