Re: st: Heteroscedasticity test for specific explanatory variable

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Mon, 13/12/10, Sun, Yan (IFPRI) wrote:
> I have an equation to estimate:
> Y=a+b1*X1+b2*X2+b3*X3+B4*X4+error term,
> Y is continuous variable, how do I conduct
> heteroscedasticity test for each explanatory variable
> (X1, X2 X3 abd X4)(in other words, I want to have a
> formal way to test if X1 or X2/X3/X4 causes the
> heteroscedasticity problem? 

I would not use a formal test for that. I suspect that
if such a test exist it will suffer from low power, so
will be much less useful than just looking at a plot of 
the residuals against each regressor.

However, if you really want to do that you could look at:

Christopher F. Baum (2006) "Stata tip 38: Testing for 
groupwise heteroskedasticity". The Stata Journal, 6(4):
590--592.
<http://www.stata-journal.com/article.html?article=st0117>

> I know there is a paper Pagan and Vella (1989) dealing
> with this, but I do not know the STATA code to do this?

This is common error on the list. When refering to a 
publication you _must_ give a full reference. Anything 
less will only waste your time and ours. This is clearly 
explained in the Statalist FAQ, a link to which can be 
found at the bottom of every post on the statalist.

-- Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/