st: Remedy for serial correlation in Panel Data

["D. Demetriou" <dd358@cam.ac.uk>]

Hi everyone,

Just a quick question: what is the best way to account for Serial 
correlation in Panel Data (T=10, N=9)?

Some (e.g. Cameron&Triverdi) say that autocorrelation is more a problem in 
long macro panels=> I shouldn't worry according to them

Others say that it is certainly a problem, irrespective of circumstances, 
since it violates one of the basic assumtpions of panel data and gives 
'wrong' standard errors => I should worry very much according to them

So?? Any ideas on this?

PS: I am using 'xtserial' test (Wooldridge test for autocorrelation in 
Panel Data) to detect whether or not serial correlation is a problem.

Yours,

DD

*
*   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/