Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Fwd: Conditional Logit estimation problem


From   S J <econhead@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Fwd: Conditional Logit estimation problem
Date   Thu, 15 Apr 2010 12:12:33 -0400

Hello,

I would like to know if anyone has ever tried to estimate the
following types of equations or knows of a way to do this in Stata.

Classical conditional logit models are run off the idea that the model
looks like:

prob(j)=exp(Bxj)/(sum exp(Bxi) for all i)

where j and i are subscripts describing the possible alternatives, B
is the parameters associated with the characteristics x. I have run
these models before using clogit.



I want to run a model like:

prob(j)=exp(G*Bxj)/(sum exp(G*Bxi) for all i)
where G is a scalar effect or function of some other characteristics
not related to the individual or to the particular dependent
variables, but on some unrelated external factor (like questionnaire
length or geographic region). I want to be able to also estimate G.



Is there any way to do this without having to write my own maximum
likelihood estimator from scratch? I am trying to find a way to run
this sort of regression without having to spend enormous amounts of
time in writing programs. Alternatively, does anyone know of a canned
program outside of Stata that can do this simply?



Thanks very much.

SJ
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index