[Limdep Nlogit List] questionnaire design with dataset and lim file.(No.2)

[William Greene]

It depends on what happened next.  If the estimator went on to find a solution
with exit code = 0, then no problem.  If it exits with a different error, then the problem
is serious.  This will be a model specification and data problem, but the way, not
a program problem.  Logit models should be easy to estimate. When they break down,
it is generally a data problem such as an attribute that is always zero or one that 
can predict an outcome perfectly.
/Bill 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: Hiroshi KAMEYAMA <kameyama.hiroshi at gmail.com>
Date: Tuesday, August 14, 2007 2:06 am
Subject: [Limdep Nlogit List] questionnaire design with dataset and lim file.(No.2)

> Dear John and to those who took care of my beginers questions,
> Yes, thank you so much for your suggestion. The problems are solved.
> 
> But the next erroe message came out.
> 
>   Hessian is not positive definite at start values.
>   Error:   803: Hessian is not positive definite at start values.
>   B0 is too far from solution for Newton method.
>   Switching to BFGS as a better solution method.
> 
> Is it seriour or negrectable?
> 
> Hiroshi KAMEYAMA
> KAGAWA Univ., Japan
> 
> 
> >This sometimes happens if you have additional rows of data 
> included or for reasons that I cannot explain. 
> When this happens to me >(often), I simply reject the rows -> 
> reject;cset=-999$ That usually solves the problem.
> >John
> 
> >>Executing NLOGIT, the following error message came out. What 
> does it 
> mean? I copy and pasted only for the dataset area again,
> there can not be other value in rows colum.
> >>Error: 1095: Count variable CSET at row 3841 equals -999
> >>Error: 1095: Count variable CSET at row 3841 equals -999
> 
> >>Row 3840 is the last row for 96 individual respondents.
> 
> 
> 
> 
> -- 
> ************************************************
> KAMEYAMA, Hiroshi (亀山　宏)
> Associate Professor
> Farm Management,  Faculty of Agriculture 
> Kagawa University, Japan 761-0795
> 2393 Ikenobe Miki-cho, Kita-gun, Kagawa 
> 〒761-0795 香川県木田郡三木町池戸2393
> 香川大学農学部　応用生物科学科生物資源生産学大講座
> Phone: +81-(0)87-891-3054 (FAX: -3021. admin.,庶務)
> URL:http://www.ag.kagawa-u.ac.jp/kameyama/index.html
> Email: kameyama at ag.kagawa-u.ac.jp
> ************************************************
> 
> _______________________________________________
> Limdep site list
> Limdep at limdep.itls.usyd.edu.au
> http://limdep.itls.usyd.edu.au
>