[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: RE: lags for xi created vars
I have created interaction between state (categorical) and income
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.
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: