I don't understand why you can't estimate a "dynamic panel model" using
xtpcse. I think by "dynamic" you mean including a lagged dependent
variable (LDV) on the right-hand side? Xtpcse is very capable of that,
and indeed the original advocates of the pcse methodology (Beck & Katz)
recommend that models using pcse should typically include an LDV.
Cheers,
Jason Yackee
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Abhijit
Sarkar
Sent: Friday, December 01, 2006 6:14 PM
To: [email protected]
Subject: st: dynamic panel with cross panel heteroscedasticity and
spatial correlation
Hi!
xttest3 and xtcsd tests after xtreg,fe show evidence of existence of
cross panel het. and spatial correlation. In a static panel setup, the
estimation can be done with xtpcse (as in my case T<<N).
My question is, if I try to fit a Dynamic panel model to the same
dataset, how can I proceed? with robust option in xtabond2, within
panel het and autocorr can be addressed, but how to specify for
cross-sectional het and correlation?
thanx in advance.
abhijit
*
* 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/
