Message-ID: <41ACE1A5.1050401@acelerate.com>
Date: 2004-11-30T21:09:57Z
From: Kjetil Halvorsen
Subject: Choice modelling (was:(no subject))
In-Reply-To: <20041130155527.8897386600@mail.fr.clara.net>
jruiz99 at ensae.org wrote:
>Hello,
>
>I am trying to estimate a choice model with varying choice set for each
>individual.
>
>I would like to fit different kinds of model (logit ,nested logit,
>probit...).
>
>So far I have found that package *mnp* allows me to estimate a probit model
>with varying choice set.
>
>But for estimation of a logit model, I have only found function *multinom*
>of package *nnet* which does not seem to allow for varying choice set.
>
>I could incorporate indicators of choice availability as explanotary
>variables, but it does not seem a very good way to do it.
>Instead, for a logit model, I have coded a likelihood computation of the
>underlying model with varying choice set and I use optim function to get
>the "maximum".
>
>
Could you post the code?
>I wondered if there were more user friendly functions available and/or
>other choice model related package.
>
>
Look at CRAN package eba (elimination by aspects).
Kjetil
>Thanks,
> Julien
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>
>
>
>
--
Kjetil Halvorsen.
Peace is the most effective weapon of mass construction.
-- Mahdi Elmandjra