st: RE: Want to run xtreg with AR(1) autocorrelation allowed

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Rung, 

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> rungporn@uchicago.edu
> Sent: 02 May 2006 18:57
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Want to run xtreg with AR(1) autocorrelation allowed
> 
> Dear All,
> 
> I want to run the panel data of 2000 entries for a 3-year 
> period.  The 2000 entries are grouped into 14 U.S. states.  
> I can use the xtreg, cluster() to capture the group effect 
> but I still need to account for the autocorrelation of the 
> error terms (AR(1)).

It depends what you mean by "accounted for".  The -cluster- option gives
you standard errors that are robust to arbitrary within-group
autocorrelation (as well as arbitrary heteroskedasticity), so in that
sense, the AC is already "accounted for".  Maybe this is enough for your
purposes?

HTH.

--Mark

> Can somebody please tell me how I 
> should do that?  I don't see whether xtreg will allow me to do AR(1).
> 
> Also, somebody suggests that I try xtgls command.  I got the 
> error message that the "Panel must be balanced."  I was 
> confused.  What does it mean by that?  Does that mean I can't 
> use xtgls?  What should I do to get both the "cluster" 
> and the "autocorrelation" aspects into my panel regression?
> 
> Thanks,
> Rung
> *
> *   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/
> 
> 

*
*   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/