Bulent <[email protected]>:
-xtgee- is an alternative; read the manual entries in [XT]. But better
IMHO is to use
xtreg, fe i(ivar) cluster(ivar)
if you have large M (number of FE/clusters) since it consistently
estimates SEs in the presence of arbitrary intrapanel correlation of
errors. (See references at http://repec.org/usug2007/crse.pdf)
On Feb 9, 2008 6:48 AM, Koksal, Bulent <[email protected]> wrote:
> Dear Stata Users,
>
> Can we use xtgls for large N small T panel data?
>
> If we have heteroskedasticity in a large N small T data, is the only
> option available to use
>
> xtreg, fe vce(robust) ?
>
> Thanks for any help.
>
> --
> B�lent K�ksal
>
*
* 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/
