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

st: Re: xttest2 doubt


From   Christopher F Baum <baum@bc.edu>
To   Hernan Vallejo <hvallejo@uniandes.edu.co>
Subject   st: Re: xttest2 doubt
Date   Mon, 17 Jun 2002 20:27:13 -0500

Prof. Vallejo,

I doubt that any results from the test have any value. Check the "obs per group" in the xtreg output. You will probably find that the min==2. The test is based
on a correlation matrix, calculated over the observations which are available for each unit. Apparently in your case there are only two time points at which all units are available. That does not prevent computation of the fe model, but the test based on a correlation matrix with two observations per unit is surely worthless.

Thanks
Kit Baum


--On Monday, June 17, 2002 17:37 -0500 Hernan Vallejo <hvallejo@uniandes.edu.co> wrote:


Dear Sir,

I have downloaded your xttest2 in order to test for heteroskedasticity in
an fixed effects panel regression.
However, when I have used it, it states

"Based on 2 complete observations"

¿What that does mean and what does that imply in terms of the validity of
the test?
Kind regards,

Hernán Vallejo
Profesor Investigador
CEDE Facultad de Economía
Universidad de los Andes
Bogotá, Colombia

*
*   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