[Limdep Nlogit List] ;asc and Information Criteria

William Greene wgreene at stern.nyu.edu
Fri Jul 20 19:57:02 EST 2007


Hendrik. You have specified a two choice model.  The two outcome probabilities are thus
functions of the difference in the utilities, which in your case is
U(vas) - U(self) = epsilon(vas) - epsilon(self).
No parameters.
/B. Greene

************************************************
Professor William Greene
Department of Economics
Stern School of Business
New York University
44 West 4th St., Rm. 7-78
New York, NY   10012
Ph. 212.998.0876
Fax. 212.995.4218
URL. http://www.stern.nyu.edu/~wgreene
Email. wgreene at stern.nyu.edu
************************************************

----- Original Message -----
From: Hendrik Ernst <hendrik.ernst at fh-gelsenkirchen.de>
Date: Thursday, July 19, 2007 10:39 am
Subject: [Limdep Nlogit List] ;asc and Information Criteria

> Dear list members,
> 
> I tried to estimate a Discrete Choice model with a Nlogit command as
> follows:
> 
> NLOGIT
> ;Lhs=wahl,anzalt,alternat
> ;Choices=vas,self
> ;asc
> ;model:
> U(vas)=price*Preis+time*zeit+quality*Qualitt+VAS*Revas/
> U(self)=price*Preis+time*zeit+quality*Qualitt+VAS*Revas$
> 
> The following error message occurred: (supposedly because of the 
> asc 
> command)
> Error  1049: Model has no free parameters among slopes!
> Error  1024: Failure estimating DISCRETE CHOICE model.
> 
> If this error is due to my "bad data" I would like to know how to 
> compute the No Coefficients Log-L Function for comparison with 
> Constants 
> Only Log-L Function and Coefficiant Log-L Function.
> 
> Additionally, since NLOGIT's output contains AIC, BIC and HQIC I 
> would 
> appreciate if anybody provides me with how NLOGIT computes these 
> criteria.
> Any help is appreciated.
> 
> Many thanks and best regards,
> 
> Hendrik.
> 
> _______________________________________________
> Limdep site list
> Limdep at limdep.itls.usyd.edu.au
> http://limdep.itls.usyd.edu.au
> 




More information about the Limdep mailing list