Hello everybody,
I have the following problem:
In a first-stage OLS regression, I predict labour income. I then create a
shock variable using predicted labour income and other, non-predicted,
variables (such as actual labour income).
This shock variable is then included as an explanatory variable in the
second stage, a difference GMM estimator which I implement using the
-xtabond- command.
How could I correct for generated regressor bias in this context? Robust
standard errors will not be sufficient, I assume. The fact that I do not
just include the predicted variable, but rather a variable which has been
generating using the predicted variable, is a further complication.
Best wishes,
David
*
* 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/
