Statalist


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

st: 2SLS with ordered logit and discrete endogenous variable


From   "Isidro, Helena" <Helena.Isidro.1@city.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: 2SLS with ordered logit and discrete endogenous variable
Date   Tue, 28 Apr 2009 09:47:41 +0100

Dear all

I am running a ordered logit model on a series of X regressors. One of
my regressor (x3) is endogeneous so I was considering a 2SLS.

The Stata command would be:

ivreg2 Y x1 x3 (x3 = z1 z2)

The problem is that Y (the dependent variavel is a ordered variable that
takes values 0, 1, 2, 3,etc.
The endogeneous regressor X3 is a discrete variable that takes values
0,1.
So the first stage would be a logit model, the second stage a ordered
probit model.

-ivprobit- assumes a continous endogenous variable (I think) and ivreg2
is for continuous variables.

Does any one has any suggestions on how to deal with this?

Thanks


Helena Isidro
Cass Business School
United Kingdom

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index