-ivreg2- will also estimate linear regression models using robust
(heteroskedastic-consistent), autocorrelation-consistent (AC) and heteroskedastic and autocorrelation-consistent (HAC) variance estimates. And, given that it is wrapper for it, -xtivreg2- do the same (only fixed effects and first differences allowed, no random effects). You can check some of your problems with -abar- (performs the Arellano-Bond test for autocorrelation). Have a look at -newey2-, also. It produces Newey-West standard errors for coefficients estimated by OLS or 2SLS. The error structure is assumed to be heteroskedastic and possibly autocorrelated up to some lag.
I think all are available from ssc.
Nicola
P.S. May-be something has gone wrong, but I recommend a more informative title to increase the likelihood to have an answer!!!
At 02.33 28/02/2007 -0500, "Nelly EXBRAYAT" wrote:
>Hi everybody,
>
>I want to use the commande xtivreg to estimate a regression in panel with
>instrumental variables.
>I will probably need to correct for serial (and maybe spatial)
>correlation. DO you know how can I test for these correlation and how
>could I correct this within xtivreg estimation?
>I woul appreciate some help.
*
* 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/
