[Limdep Nlogit List] Error in Gamma estimation

[William Greene]

Anil. The message would suggest that your data are not consistent
with the gamma model. I would not say you can ignore the message,
since the estimator barely left the starting values. This is an
indicator that this is the wrong model. There is no way to "correct"
it, save to consider a different model. The gamma model is pretty
difficult in any event. Perhaps a different specification, such as
the NegBin-P model would work.
/Bill Greene

----- Original Message -----
From: "Anil Rupasingha" <anilrup at gmail.com>
To: limdep at limdep.itls.usyd.edu.au
Sent: Tuesday, October 7, 2008 12:38:26 PM GMT -05:00 US/Canada Eastern
Subject: [Limdep Nlogit List] Error in Gamma estimation

Dear All:
I am getting following error message when estimating Gamma model with
Poisson command;

"Initial iterations cannot improve function.Status=3

Error 805: Initial iterations cannot improve function.Status=3

Function= .14813758483D+04, at entry, .10635851050D+05 at exit"
Can I ignore this or if not how do I correct it?  Thank you for any replies.

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