Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: "flat region resulting in a missing likelihood" (error message)


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: "flat region resulting in a missing likelihood" (error message)
Date   Wed, 25 Feb 2009 14:23:23 -0600

This is likely an (empirical) underidentification issue. Your -taxe-
variable probably does not vary within -country-, and limited
dependent variable models do not like it when you can predict some
outcomes perfectly from the values of the explanatory variable.

In fact, I agree with Martin, that's a weird statement of the problem
to me. -country- makes a poor dependent variable unless you are
talking about some sort of empirical IO where a big international firm
chooses where to place their headquarters... I cannot think of any
other example. If that is indeed your research question, then you are
basically out of luck. You can try to add some other explanatory
variables in the RHS of your model, but it may not work in the end,
with the parameters for -taxe- still not estimated properly (or
estimated with huge or missing standard errors).

On 2/25/09, Isabelle.Roux@ehess.fr <Isabelle.Roux@ehess.fr> wrote:
> Hello,
>
>  I'm trying to do a multinomial logit and I write to stata: "mlogit country
>  taxe" where country is the country's choice of firms and taxe is corporate
>  tax of the country.
>
>  Stata refuses to provide results and indicates: "flat region resulting in
>  a missing likelihood"
>
>
>  In Stata's help I have this message:"convergence not achieved;
>         You have estimated a maximum-likelihood model and Stata's
>         maximization procedure failed to converge to a solution;
>         see [R] maximize.  Check if the model is identified."
>
>  I don't know if I have a problem of stata's command or of econometrics; in
>  this last case, could you indicate me please which topic to study to solve
>  my problem?
>
>
>  (maybe it is a very obvious question but I am new in econometrics and I'm
>  a bit lost in regressions!)
>
>  Thank you very much for help,
>
>  Isabelle
>
>  *
>  *   For searches and help try:
>  *   http://www.stata.com/help.cgi?search
>  *   http://www.stata.com/support/statalist/faq
>  *   http://www.ats.ucla.edu/stat/stata/
>


-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index