Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Poisson regression -not concave-


From   "Michael N. Mitchell" <Michael.Norman.Mitchell@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Poisson regression -not concave-
Date   Sun, 06 Jun 2010 21:15:43 -0700

Greetings

I would suggest that you try making a simple crosstab of the "dummy" variable by the dependent variable. I think that will show that the outcome variable cannot be zero when the dummy variable is 1. It is possible that the estimation command is "bothered" by this determinacy between the predictor and outcome, creating a situation where the procedure cannot converge. I am sure that others could provide a more technical and detailed explanation, but I hope this is helpful.

Best regards,

Michael N. Mitchell
Data Management Using Stata      - http://www.stata.com/bookstore/dmus.html
A Visual Guide to Stata Graphics - http://www.stata.com/bookstore/vgsg.html
Stata tidbit of the week         - http://www.MichaelNormanMitchell.com



On 2010-06-06 9.07 PM, Paolina Medina wrote:
Stata listers,

I'm trying to run a poisson regression model in which the dependent
variable is the number of mobile phones from the households, and among
other explanatory variables i am using a dummy wariables which takes
the value of 1 when the household has at least one cell phone and zero
otherwise.

But it turns out that whenever i use that dummy variable the iterative
process in stata does not converge, and it can be read for example:
iteration 380: log likelihood = -7734.6891  (not concave)

It i drop that dummy variable i get convergen at Iteration 3:   log
likelihood = -7734.6927

Does anyone has any suggestion or explanation?
And where can i get more information about the numerical process
behind the maximum likelihood estimation?

Thank you very much in advance.

PM
*
*   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/
*
*   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   |   Site index