[Limdep Nlogit List] Kernel Densit Estimator/Log likelihood ratio test

[William Greene]

Mr. Dutta.
(1) I would say probably no. The KDE is not an optimum of any
sort, so it can't distinguish models.
(2) In some cases, the Vuong statistic can be used. It's written
up in my text.
/B. Greene
----- Original Message -----
From: "Santa Dutta" <send2santa83 at gmail.com>
To: limdep at limdep.itls.usyd.edu.au
Sent: Wednesday, February 20, 2008 11:25:11 PM (GMT-0500) America/New_York
Subject: [Limdep Nlogit List] Kernel Densit Estimator/Log likelihood ratio test

Sir,

I have the following doubts while working with NLOGIT(3.0):

1)Can the kernel density estimator be used as a guide to identify the type
of distribution suitable for a particular parameter while estimating Random
Parameter logit Models?

2)Log likelihood cannot be used as a basis for comparing two models with
same number of parameter estimates(degrees of freedom).
Can you please suggest some test which can be used for model
selection/identification in this case.

Can NLOGIT 4.0 help me in any way regarding these?

Regards,

S.S.Dutta
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au


-- 
Professor William Greene
Department of Economics
Stern School of Business
New York University
44 West 4th St., Rm. 7-78
New York, NY   10012
http://www.stern.nyu.edu/~wgreene
212.998.0876