Dear Stata List:
I have a panel data of 51 states over 26 years.
I understand that if I suspect that there is within panel residual
correlation, then I have use the cluster option with -xtreg, fe- to
correct for the correlation factor in my SEs.
I suspect that such a situation might be true for my data. But when I
run Wooldbridge's -xtserial- test after -xtreg, fe- it says that there
is no auto-correlation in my residuals.
Does this mean that I should not use cluster option with -xtreg, fe- ?
many thanks for your help
amin
*
* 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/
Locked
