Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Maximum sizes Nested Logit

From   <>
To   <>
Subject   RE: st: Maximum sizes Nested Logit
Date   Wed, 16 Aug 2006 15:04:48 +0100

Just a few clarifications to the answers I have received from Anders
Alexandersson and Rich Gates:

I am using nlogit:

nlogit choice (n2_code = n2_gdp n2_mp n2_w...) (c_code = c_gdp c_mp
c_w...), group(id)

I have 11 explanatory variables in the upper level and 9 in the lower.
Alternatives in the lower level=246 (regions)
Alternatives in the upper level=25 (countries)
Therefore, it is a two level nested logit.
Observations: 6,475 firms that decide to locate in those regions
belonging to those countries (they are not alternatives, I didn't
explain it clearly in my first email, although I guess that there are
still a lot of parameters to estimate, but "only" 25 inclusive values)

I can estimate simpler nested models, eg. Taking only 3 countries and
their regions.

I guess it is a problem of time since the large number of parameters
that are required to be estimated as Rich pointed out and have to find a
more powerful computer ??? if stata doesn't have any size limit in the
number of alternatives and nests.

Thanks for your comments and I really appreciate any other consideration
you may have,


-----Original Message-----
[] On Behalf Of Anders
Sent: 16 August 2006 14:16
Subject: Re: st: Maximum sizes Nested Logit


Which command are you using for the nested logit: gllamm, nlogit, or
nlogitrum? Exactly which syntax did you use? Can you estimate simpler
nested logit models, e.g., the models in [R] nlogit? If you -set trace
on- do you see any output?

Anders Alexandersson

Rodrigo <> wrote:

> I am trying to estimate a nested logit model with the following
> characteristics:
> -       Location choice firm in 25 countries (upper level) and 246
> regions belonging to those countries (lower level)
> -       I have 6,475 location choices.
> -       Explanatory variables are economic country and region
> characteristics.
> The estimation does not work. It recognizes the nested structure but
> after that the computer seem to run the model but not results appear,
> even no iterations.
> I was wondering whether there is a maximum size restriction in the
> number of alternatives, nests or observations to estimate a nested
> with Stata.
*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index