Thanks for your answer Kit,
however, How should I test the autocorrelation within the panel in my panel
dataset (T=31 and N=143)? Should I use xttest2 just for a subsample?
thanks Giorgio
----- Original Message -----
From: "Christopher F Baum" <[email protected]>
To: <[email protected]>
Sent: Friday, October 17, 2003 1:10 PM
Subject: st: Re: autocorrelation in fixed effects context
> On Friday, October 17, 2003, at 02:33 AM, Giovanni wrote:
>
> >
> > 5) I've tried to test the autocorrelation by the command xttest2 but I
> > have alwaiys the same error: the correlation matrix is singular. What
> > does it really mean?
>
> In correspondence off-list, Giovanni indicated that the number of units
> in his balanced panel far exceeds the number of time-series
> observations per individual. Since my routine xttest2 works with the
> same correlations which would be used in the application of SUR (sureg)
> to panel data in 'wide' format, it will report a singular correlation
> matrix whenever T < N (and in that context sureg will not work). You
> cannot have N linearly independent vectors of length T if T<N, as it
> does in his context.
>
> Kit
>
> *
> * 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/
