You don't specify under # 1 whether you are running regular OLS or the Beck & Katz (1995) flavor, which is implemented as - xtpcse -. If you haven't tried the latter, see if that does anything for you, though there is some debate over whether B&K is better than a fixed effects GLS estimator with robust standard errors given certain conditions. See http://polmeth.wustl.edu/retrieve.php?id=54#search=%22%22panel%20corrected%20standard%20errors%22%22
Jason Webb Yackee
Fellow
Gould School of Law
University of Southern California
Los Angeles, California 90089-0071
Tel: 213-740-2865
Fax: 213-740-5502
Cell: 919-358-3040
email: [email protected]
�
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of SHRUTI KHADKA MISHRA
Sent: Monday, September 18, 2006 8:47 AM
To: [email protected]
Subject: st: request for suggestions on fixing heteroscedasticity in fixed effect model
Hi,
My question is how can we fix heteroscedasticity problem in a Fixed effect model?
I am running a panel model of 26 countries and 24 years.
I found two solutions in STATA to deal with heteroscedasticity.
1. OLS with dummies to make it fixed effect and run Robust Error to correct heteroscedasticity. But then, I have many varibles which are already dummies themselves, so its hard to deal with collinearity problem.
2. xtgls. But I can not relate if and how to turn FGLS into a fixed effect model.
Do you guys have better ideas to deal with it?
Appreciate your help.
Shruti Khadka Mishra
Graduate Student
The Ohio State University
*
* 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/
