Statalist


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

Re: st: question about subdue multiple dummy variables


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: question about subdue multiple dummy variables
Date   Tue, 13 May 2008 09:41:32 -0400

Or just:
egen long i=group(d1,d2)
areg y x, absorb(i)

On Tue, May 13, 2008 at 9:10 AM, Rodrigo Alfaro A. <ralfaro@bcentral.cl> wrote:
> Well, let's suppose that c1 and c2 are your categorical variables.
>  Moreover, c1={1,2,3} and c4={1,2,3,4}... then I will try something like:
>  gen double c3= c1*10 + c2, then areg whatever, a(c3).

On Tue, May 13, 2008 at 9:07 AM, Joseph Coveney <jcoveney@bigplanet.com> wrote:
> Ziyang Wang wrote:
>
>  I'd like to ask a question about dummy variables, I have 11 dummy variables
>  which result from the interaction of two dummies, one with 3 categories and
>  the other with 4 categories. I like to know how to subdue those dummies in
>  the regression output? Something like absorb option in areg. Anyone knows?
>  Many thanks!
>
>
> --------------------------------------------------------------------------------
>
>  If it's a linear model, then you can use -anova-.  Otherwise, you can't
>  suppress the reporting of individual levels of dummy variables as far as I
>  know, but you can use -test- afterwards to get a single test statistic.
>
>  Joseph Coveney
*
*   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