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. <[email protected]> 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 <[email protected]> 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/
