Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Price-Winston regression for panel data - xtpcse


From   "bondaanton" <bondaanton@yandex.ru>
To   statalist@hsphsun2.harvard.edu
Subject   st: Price-Winston regression for panel data - xtpcse
Date   Tue, 15 Nov 2005 16:00:45 +0300 (MSK)

I have problems with heteroscedasticity in my panel.
Then,I can't use GLS to correct for it,as my panel has T<N.
I can use  robust version of areg or Price_Winston regression for panel data. I cannot choose between them, as Stata does not provide any information about assumptions of these two methods. Even worse, if I choose,say, P-W regression, I also have to choose type of correlation within groups and heteroscedasticity, Tests show I have all kinds of them. So, if not understanding how this method works, I cannot secure myself against low degrees of freedom. The main question is: how the variance-covariance matrix is estimated in the franework of P-W regression? Is it possible, say, estimate panel of 214x6=1284 observations with 10 coefficients, AR(1) within correlation, contemprorary correlation and heteroscedasticity? I am not sure in this, as covariance matrix has 23000 elements... 

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index