Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Re: testing for heteroskedasticity


From   Kit Baum <baum@bc.edu>
To   "Alessandro Giambattista" <a.giambattista@gmail.com>
Subject   st: Re: testing for heteroskedasticity
Date   Thu, 30 Oct 2008 06:02:45 -0400

Don't bother. Do the Breusch-Pagan test (estat hettest in official Stata) using the single variable on which you would divide the data into subsamples (e.g., firm size) in its varlist. The G-Q test is outdated and has very low power relative to the B-P test, as many textbooks will tell you.

Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Oct 29, 2008, at 22:01 , Alessandro Giambattista wrote:

Hi, I'm not sure I'll get an answer, but still it's worth to try:

I'm trying to do the Goldfeld-Quant test for heteroscedasticity with Stata but I do not know hot to perform it. I've been told I have to divide the sample in two subsamples, somehow sort the data and then perform it.

Could you help me?

Thanks,

An econometric student

*
*   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/



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