st: multinomial probit/logit model with endogenous variables

[xueliansharon <xuelianstata@gmail.com>]

Dear all,

I want to estimate a multinomial probit/logit model, but I have 1 endogenous
variable in the model, and I have 2 IVs in hand. Is there anyway to address
the endogeneity problem? Or maybe I should use two-stage multinomial
probit/logit? 

i.e.
1st-stage
regress y x1 x2 x3 I1 I2
predict yhat, xb

2nd-stage
mprobit z yhat x1 x2 x3

Is this method correct? (here, z is the multinomial dependent variable, y is
the endogenous variable, x1, x2, x3 are the exogenous variables, I1 and I2
are instruments)

Any help will be appreciated!

Sharon


--
View this message in context: http://statalist.1588530.n2.nabble.com/multinomial-probit-logit-model-with-endogenous-variables-tp6366748p6366748.html
Sent from the Statalist mailing list archive at Nabble.com.
*
*   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/