[Limdep Nlogit List] help with error 590: logminus

William Greene wgreene at stern.nyu.edu
Sat Oct 13 07:44:42 EST 2007


Christy.  The log() function is being fed a nonpositive value. (It might
be zero.)  That is the source of the problem.
/B. Greene
----- Original Message -----
From: loper at UDel.Edu
To: limdep at limdep.itls.usyd.edu.au
Sent: Friday, October 12, 2007 4:01:08 PM (GMT-0500) America/New_York
Subject: [Limdep Nlogit List] help with error 590: logminus

Dear all,

I am trying to estimate a couple of different double bounded dichotomous choice models and keep getting the same error message for both models.  Has anyone received this before?

Error   590: Obs.=     1 Cannot compute function: Logminus

I know there is not an issue with the first observation.  Perhaps it is trying to take the log of a negative number?

The code that I am using for the simpler model is below; it was adapted from a model that Alessandro Corsi posted to the listserv in 2006.  

Any suggestions or guidance are welcome!


--> create;
   yy= (i1=1)*(i2=1);
   ny= (i1=0)*(i2=1);
   yn= (i1=1)*(i2=0);
   nn= (i1=0)*(i2=0)$
--> maximize; labels=b1,b2; start=0,0;  

[note: I get the error no matter what starting values I use]
[a= bid 1, ah= higher bid, al=lower bid]
  
;fcn=
   xxh= -b1+b2*ah |
   pyy=log(phi(-xxh))  |
   xxx=-b1+b2*a  |
   pyn=log(phi(xxh)-phi(xxx))  |
   xxl=-b1+b2*al  |
   pny=log(phi(xxx)-phi(xxl))  |
   pnn=log(phi(xxl))  |
   yy*pyy+pyn*yn+pny*ny+pnn*nn$

Cheers,
Christy Loper
Ph.D. Candidate
University of Delaware
College of Marine and Earth Studies 
_______________________________________________
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




More information about the Limdep mailing list