Would estimating your model using -reg, cluster - followed by the
Arellano-Bond test for serial autocorrelation -abar- accomplish anything
useful in this regard? This works for unbalanced panels, but I don't
know if what it gives you is equivalent to what you are trying to get
out of -xtserial, panels(correlated)-.
JY
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Christian
Andres
Sent: Friday, September 08, 2006 3:01 AM
To: [email protected]
Subject: st: Autocorrelation in unbalanced panels
Dear all,
I tested for autocorrelation in a panel dataset using xtserial as
described in http://www.stata.com/support/faqs/stat/panel.html. The
results indicate that the null of no serial correlation can be rejected.
As a consequence, I would like to estimate the model using the option
panels(correlated) in order to get consistent standard errors. However,
it seems like I can only use the option with a balanced panel!
How can I apply it to an unbalanced panel? As far as I understand
Wooldridge, the GLS transformation that eliminates serial correlation in
the errors can be extended to unbalanced panels!?!
Thanks a lot!
Christian
*
* 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/
