[R] Mixed conditional logit model

[Ryuichi Tamura]

Hi,

Sven Müller <sven.mueller <at> tu-dresden.de> writes:

> i wonder whether it is possible to estimate a mixed (random parameters) 
> logit model in R. 

I wrote a collection of R functions for estimating discrete choice models 
by simulated maximum likelihood. It includes:
- likelihood and gradient functions for estimating mixed mnl, mixed 
   panel mnl with some specific random structures for coefs
  (normal, lognormal, triangular, uniform)
- some IIA testing functions includes 'artificial variable test' in
   Mcfadden and Train(2001)
- functions for generating halton sequences for 'efficient' simulation.

Please email me if you interested in.
However, it is not so difficult to write programs for your own needs.
URL for main reference is Kenneth Train's Home Page:
http://elsa.berkeley.edu/~train/
My R functions is based on GAUSS program listed in this site.


Regards,

Ryuichi Tamura