Dear all,
for those of you who might be interested. As of now Stata doesn't do Mixed
Logits but I hope it will be incorporated into future versions. For now, if
anybody is interested, there are three venues (according to my research) as
I explain ahead. In general, it seems to me that none of the options are
more user-friendly than Stata. I do hope Stata will develop mixed logits.
I'm wondering if Stata couldn't create a command by "translating" Dr.
Train's code from Gauss?
Anyway, the options are:
1 ) Limdep (and Nlogit),
2) Gauss code of Prof. Train at UC Berkeley
3) Biogeme (free software) - requires running commands from DOS.
Option 2) seems to be the most powerful but requires Gauss.
Sincerely,
Julia
Integrated Program on Urban, Regional and Global Air Pollution
MIT
----- Original Message -----
From: "Julia A. Gamas" <[email protected]>
To: <[email protected]>
Sent: Wednesday, May 05, 2004 4:04 PM
Subject: Mixed logit?
>
> I have two questions that I hope someone can answer for me:
> I need to do estimation using a multinomial logit model that allows for
the
> error term to be flexible; mixed logit.
> 1) Does Stata do this type of estimation?
> 2) If so, which comand should I use? Is xtlogit what I need?
> Any help will be greatly appreciated.
> Sincerely,
> Julia A. Gamas
> Integrated Program on Urban, Regional and Global Air Pollution
> MIT
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
