Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: single p-value for categorical variables and interactions


From   "Clive Nicholas" <Clive.Nicholas@newcastle.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: single p-value for categorical variables and interactions
Date   Wed, 15 Nov 2006 21:48:30 -0000 (GMT)

Maarten Buis replied to Kelly Richardson:

> yes, with the -test- command, see example:
>
> *------------ begin example --------------
> sysuse auto, clear
> recode rep78 1/2=3
> xi: reg price i.foreign*i.rep78
> test _Irep78_4 = _Irep78_5 = 0
> test _IforXrep_1_4 = _IforXrep_1_5 = 0
> *------------- end example ---------------

Yes, but don't forget that you need to include _all_ the constituent
variables along with the interactions in order to obtain both the correct
parameter values and the p-values, otherwise you assume -foreign- and
-rep78- to equal 0: and that's almost never the case. So the above model
should be:

. xi: reg price foreign rep78 i.foreign*i.rep78

Admittedly, the model estimates don't change much, but some of the -test-
results do. Compare:

. test _Irep78_4 = _Irep78_5 = 0

 ( 1)  _Irep78_4 - _Irep78_5 = 0
 ( 2)  _Irep78_4 = 0

       F(  2,    63) =    0.55
            Prob > F =    0.5803

without the constituent variables, with

. test _Irep78_4 = _Irep78_5 = 0

 ( 1)  _Irep78_4 - _Irep78_5 = 0
 ( 2)  _Irep78_4 = 0
       Constraint 2 dropped

       F(  1,    63) =    0.17
            Prob > F =    0.6847

when we include them. OK, it's a silly dataset, but the point is an
important one!

CLIVE NICHOLAS        |t:
Politics              |e: clive.nicholas@ncl.ac.uk
Newcastle University  |http://www.ncl.ac.uk/geps

Whereever you go and whatever you do, just remember this. No matter how
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
conditions.

*
*   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