Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: multinomial probit/logit model with endogenous variables

From   xueliansharon <>
Subject   st: multinomial probit/logit model with endogenous variables
Date   Sun, 15 May 2011 15:14:41 -0700 (PDT)

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

regress y x1 x2 x3 I1 I2
predict yhat, xb

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!


View this message in context:
Sent from the Statalist mailing list archive at
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index