--- [email protected] wrote:
> According to stata/help, the command --test-- perform a Wald test, but
> outputs the F-statistic rather than the (related) Wald-statistic. Is there
> a way to fix that? (and if not, why?)
>
> I have also noticed a difference between the output of --test-- following a
> call to --reg-- and a call to --logit--. A call to --logit-- outputs a chi2
> (the Wald-statistic) while a call to --reg-- outputs an F.
>
> Is there a simple way to output chi2 after a call to --reg-- and --test-- ?
The F-statistic allows you to take into account the fact that you are
estimating your statistic in a finite sample. The Chi square statistic is
an approximation which assumes an infinite sample. Whenever it is possible
Stata will use the more appropriate F-statistic. This is possible in an
OLS regression like the one performed by -regress- and it is not possible
after a maximum likelihood estimation like -logit-.
You can trick Stata into thinking that it is not possible to take into
account that the sample is finite by estimating the model using -glm-
instead of for instance -regress-, as is shown in the example below.
(In essence you are estimating an OLS regression using maximum
likelihood.)
*-------- begin example -----------
sysuse auto, clear
reg price mpg foreign
test foreign
glm price mpg foreign, family(gaus) link(identity)
test foreign
*--------- end example ------------
(For more on how to use examples I sent to the Statalist, see
http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
