Johan,
thank you very much for the hints. This was exactly what I thought, too. Since xtserial is no postestimation command, it should have enough power for the panels I have, no matter what kind of regression I use. I'll also try abar.
Best
Tobias
________________________________________
Von: [email protected] [[email protected]] im Auftrag von Johan Hellstrom [[email protected]]
Gesendet: Freitag, 10. Juli 2009 10:15
An: [email protected]
Betreff: Re: st: Autocorrelation test Pooled OLS
Tobias,
Although your data have a small-T and large-N set-up, it is still a panel.
'xtserial' should probably work in your case (as the Wooldridge's test
should have descent power given your sample size), but you can also try
another user-written program - 'abar' (type 'findit abar').
Best,
Johan Hellström
*
* 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/
*
* 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/
