Re: st: RE: Goldfeldt-Quant versus -hettest-

[Ernest Berkhout <ernestb@seo.fee.uva.nl>]

At 04:54 04/02/2004, Richard Williams wrote:
> Also, some of the examples did show how to use Stata to do the 
> calculations for you, but I think this is one of those cases where it may 
> be easier just to do it on a calculator.
Effectively, I find it not so difficult to let Stata do the calculations. 
Advantage is that it does it for several groups and several variables 
within a foreach-loop. The trick is to refer directly to the scalars e(rss) 
and e(df_r). A possible setup to perform the GQ-test for a variable named 
"ophoogf2" would be:
*************************************************************************
egen group = cut(ophoogf2), group(3)
tabstat ophoogf2 , s(mean min max) by(group)

foreach group of num 0 2     {
        quiet xi: reg lnloon $rhsvars_xi if group==`group'
        scalar ee_oph_`group' = e(rss)
        scalar df_oph_`group' = e(df_r)
}
scalar GQ_oph = max(ee_oph_0,ee_oph_2)/min(ee_oph_0,ee_oph_2)
scalar dir GQ*
*************************************************************************

But then, I still have to agree with Nick Cox' earlier argument that only 
looking at the p-value is not enough, one might first take a look at 
residuals & outliers.


Ernest Berkhout
SEO Amsterdam Economics
University of Amsterdam

Room 3.08
Roetersstraat 29
1018 WB Amsterdam
The Netherlands

tel.:+ 31 20 525 1657
fax:+ 31 20 525 1686
http://www.seo.nl
===========================
A statistician: someone who insists
on being certain about uncertainty
===========================

*
*   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/