[Limdep Nlogit List] (no subject)

[William Greene]

That kind of message often signals that you have the wrong model. In this case,
evidently there is no evidence of random effects contained in the data. If you 
suspect the specification of the RHS, try fitting the model with only one or a
very small number of variables, and build it up a little at a time (assuming it
gets started at all, which it may not).  Note that this has nothing to do with
the algorithm. The RE truncation model uses BFGS.  It's not possible that one of
the other algorithms will improve on this.
/B. Greene



----- Original Message -----
From: "Craig Gallet" <cgallet at sbcglobal.net>
To: limdep at limdep.itls.usyd.edu.au
Sent: Friday, August 29, 2008 5:35:20 PM GMT -05:00 US/Canada Eastern
Subject: [Limdep Nlogit List] (no subject)

Hello:
 
I am currently trying to estimate a truncated model (lower tail limit for lhs variable = 0) using random effects.  I continue to get the following error messages:
 
Warning   137: Iterations: function not computable at crnt. trial estimate
  Error   802: Cannot compute function at current values. Exit status=4
All of the rhs variables in my model are dummy variables.  Is this the likely reason for these error messages?  I'm able to estimate the model without truncation.  Also, if I change the lower limit to say -0.5, I can get the model to run (but nontheless need 0 for the lower limit).  I've also tried alternative algorithims, but they have failed each time.  Any suggestions would be greatly appreciated.
 
Craig
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au