Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: xtabond + endogeneity


From   "Cordula Stolberg" <[email protected]>
To   [email protected]
Subject   st: xtabond + endogeneity
Date   Sun, 19 Dec 2004 12:54:04 +0100 (MET)

Hello,

I am estimating a dynamic panel with xtabond. I suspect that one regressor
might be endogenous. In earlier postings, I have read that it is possible to
use a Hausman test to test for the endogeneity of regressors. However, I am
not quite sure about the syntax.

Assuming I have the semi-log model

logY = logY(t-1) X1 X2 TimeDummies

and I suspect X1 to be endogenous. I know that I can estimate the model
treating X1 first as endogenous (always consistent) and then a second model 
treating X1 as exogenous (consistent only if X1 is exogenous) and do a
-hausman model1 model2-. 

However, I am not sure about the syntax. With -xtabond-, would that look
like:

Model 1 (always consistent):

xi: xtabond logY X2 i.TimeDummies, lags(1) pre(X1, lags(1,.) end)

Model 2 (consistent only if X1 is exogenous):

xi: xtabond logY X1 l.X1 X2 i.TimeDummies, lags(1)


Is this the correct syntax? And with -xtabond2- would this be expressed as:

Model 1:

xtabond2 logY l.logY l(0/1).X1 X2 TimeDummies, gmm(l.logY, X1) ivstyle(X2
TimeDummies)


Model 2:

xtabond2 logY l.logY X1 l.X1 X2 TimeDummies, gmm(l.logY) ivstyle(X1 l.X1 X2
TimeDummies)


Thanks for your help,
Cordula
*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index