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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: mlogit iterations not concave


From   Jessica Bishop-Royse <jessibishoproyse@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: mlogit iterations not concave
Date   Sun, 21 Mar 2010 22:42:26 -0400

Partha,
Thank you for the iter(20) option.  I did seem to find some of my
items that were problematic.  Not sure there is anything I can do
about it, since I feel like I should include the variables that were a
problem (maybe reconsider what I am using as a reference group).  I
appreciate your help.

Thanks.

jcbr

On Sun, Mar 21, 2010 at 4:46 PM, Partha Deb <partha.deb@hunter.cuny.edu> wrote:
> Jessica,
>
> I suggested you reduce the number of categories for your dependent variable.
>  It should help you identify sources of singularity or near-singularity,
> which is almost certainly the reason for the "non concavity".  I though you
> were including county-indicators but now it seems that you have county-level
> continuous variables.  If so, keep them as continuous variables.
>  Categorizing those makes is more likely you run into problems with "empty"
> cells.  Your overall problem is, almost surely, that your model is not well
> specified.
>
> To get a feel for where your problem is add the -iter(20) option to your
> -mlogit- command.  -mlogit- will then stop iterating after 20 iterations and
> produce a table of coefficients / standard errors.  Look for coefficients
> that are too big / small and/or standard errors that are too big / small or
> "." .
>
> Hope this helps.
>
> Partha
>
>
> Jessica Bishop-Royse wrote:
>>
>> Dear StataListers,
>> I am working on a project where I am trying to estimate a mlogit for a 9
>> category nominal dv (cause of death), using both county level predictors
>> (67
>> counties) and individual level predictors (continuous vars like education
>> and age as well as dummied vars like not married). When I run county level
>> ivs by themselves with the dv, I get significant results.  When I run
>> individual level ivs by themselves with the dv I get signficant results.
>> When I try to incorporate both, county level predictors don't seem to be
>> significant.
>>
>> I originally was trying to estimate multilevel models (gllamm) when it was
>> suggested that I try mlogit for the timebeing.  I also dummied the county
>> level variables (so that they are no longer continuious, but rather
>> representative of established cutpoints).  It was also suggested last week
>> that I collapse my dv into fewer categories.  However, when I do this,
>> Stata
>> runs and runs and runs and gives me the message that iterations are "not
>> concave".  However, no such problems modeling with the 9 category
>> dependent
>> variable (just not significant results for county level predictors).
>>
>> My question is this:  what is this "not concave" message?  And is this an
>> indicator that I need to abandon mlogit? Opinion?
>>
>> . mlogit cause6death black maternaled maternalage real_parity for_born
>> pncint pncadqp pncindq crisisbirth multiple if birth1980==1, rrr
>>
>> Iteration 0:   log likelihood =  -10390.13
>> Iteration 1:   log likelihood = -10242.229
>> Iteration 2:   log likelihood = -9337.9062
>> Iteration 3:   log likelihood = -9238.5535
>> Iteration 4:   log likelihood = -9220.1816
>> Iteration 5:   log likelihood = -9217.2186
>> Iteration 6:   log likelihood = -9216.5682
>> Iteration 7:   log likelihood = -9216.4613
>> Iteration 8:   log likelihood = -9216.4493
>> Iteration 9:   log likelihood = -9216.4468
>> Iteration 10:  log likelihood = -9216.4463
>> Iteration 11:  log likelihood = -9216.4461
>> Iteration 12:  log likelihood = -9216.4461
>> Iteration 13:  log likelihood = -9216.4461  (not concave)
>> Iteration 14:  log likelihood = -9216.4461  (not concave)
>> Iteration 15:  log likelihood = -9216.4461  (not concave)
>> Iteration 16:  log likelihood = -9216.4461  (not concave)
>> Iteration 17:  log likelihood = -9216.4461  (not concave)
>> --Break--
>> r(1);
>>
>> end of do-file
>>
>> --Break--
>> r(1);
>>
>> --
>> Jessi Bishop-Royse
>>
>> http://sites.google.com/site/wakullagirlssoccer/
>> http://sites.google.com/site/jessicacbishoproyse/Home
>>
>> "Without a struggle, there can be no progress."
>> Fredrick Douglass
>>
>> *
>> *   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/
>>
>
> --
> Partha Deb
> Professor of Economics
> Hunter College
> ph:  (212) 772-5435
> fax: (212) 772-5398
> http://urban.hunter.cuny.edu/~deb/
>
> Emancipate yourselves from mental slavery
> None but ourselves can free our minds.
>        - Bob Marley
>
> *
> *   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/
>



-- 
Jessi Bishop-Royse

http://sites.google.com/site/wakullagirlssoccer/
http://sites.google.com/site/jessicacbishoproyse/Home

"Without a struggle, there can be no progress."
Fredrick Douglass

*
*   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