Re: st: RE: lags for xi created vars

[Ed Blackburne <blackburne@shsu.edu>]

Stas,

I have created interaction between state (categorical) and income 
(continuous) variables:

xi i.st*inc for example.

I want to estimate a different slope for each state in a regression and 
was playing around with xi to make sure I could do it.

-Ed



Stas Kolenikov wrote:


> I am just curious -- why would you need to take lags of anything
> produced by -xi-? I usually think of the categorical variables as
> something fixed, like one's race or gender.
>
>  
>
> > > > I have created a set of vars using xi command that has resulted in
> > > > _IstXab_1 _IstXab_2, etc.
> > > >
> > > > The following reference to the varlist works:
> > > >
> > > > regress y _IstXab*
> > > >
> > > > However, I cannot use lag or difference operators:
> > > >
> > > > regress y l._IstXab*
> > > >
> > > > results in error of:
> > > >
> > > > _IstXab* ambiguous abbreviation
> > > >        
>  
*
*   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/