[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Nested logit with shares/grouped data
I'm not an expert here, but I think the experts,
whoever they are, would want more than this. In particular,
the report of "different results" looks impossible
to discuss without seeing them. You may be better
off taking a dataset, getting results from Stata and
results from Limdep and mailing them both to Stata
> How do you estimate a nested logit model when your left hand
> side variable takes the form of a count (or a market share).
> i.e. the dataset records how many sales of each product are
> made in each time period.
> The stata web site offers advice for a logit or multinomial
> logit model:
> This advice suggests first putting your data in "long" form
> and then use frequency weights (fweights) with the logistic,
> logit, or mlogit command.
> I guess my question is, can such a procedure be used in the
> case of a nested logit model? I tried this procedure and got
> some answers out that look sensible. However when I checked
> it with LIMDEP NLOGIT (which claims to be able to cope with
> shares/grouped data) I get different results.
> Is this because you cannot proceed as I suggest above (or
> because LIMDEP is wrong)?
* For searches and help try: