< >
Rather than explicitly estimating the AR structure of the errors
(which is difficult in any case beyond AR(1)), use a HAC
(heteroskedasticity- and autocorrelation-consistent) estimator, a la
"Newey-West". You can do this with -ivreg2- (from SSC) or -ivregress-.
Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html
On Oct 17, 2008, at 02:33 , Gaurav wrote:
I am trying to estimate a model with AR disturbances but which
contains some
endogenous variables that I want to instrument. In other words,
what I am
looking for IV estimation with AR error terms. ivreg does not seem
to allow for
AR errors and arima does not allow for instrumental variables.
How do I do that in Stata?
*
* 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/
