[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: iv probit and endogenous binary variables
Your question I see as a statistics question,
modulo your use of econometric terminology, "endogenous"
and (tacitly) "instrumental variables". I am not aware of
anything in your question that is intrinsically econometric.
The list is declared in the FAQ to allow statistics
questions. Many people would rather that others didn't
allow questions that are very elementary or that are
easily answered by consulting standard texts, but
yours doesn't appear to be either, so no problem there.
However, there is general advice, often ignored,
"Many questions will be of interest to only some Statalist
members. Always recall that members come from many different
sciences. Try to show a little sensitivity to those
noneconometricians, nonbiostatisticians, or whoever it may
be who may know little about your question. Flag your
question as of limited interest, so that people can
delete it quickly, or take the trouble to avoid
That's a counsel of perfection, and no one is suggesting,
say, that anybody has to include a mini-tutorial on
instrumental variables for the sake of those who
are in blissful ignorance.
That said, it is always a little puzzling when people
send questions that they themselves appear to think are
inappropriate for the list.
> What should one do if one also has an interaction term with
> this endogenous binary variable on the RHS? (I realise
> this has little to do with Stata and is simply an
> econometrics question, so please excuse me for that.)
* For searches and help try: