Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: AW: st: maximum number of outcomes in mlogit


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: AW: st: maximum number of outcomes in mlogit
Date   Wed, 21 May 2008 13:38:18 +0100

I don't see why! No-one has made a convincing case for models with that
many categories. 

Richard Williams

At 05:38 AM 5/21/2008, Nick Cox wrote:
>I have no inside knowledge on this detail, but it wouldn't surprise me
>if that limit was based on simulations showing how far -mlogit- can
>deliver credible and intelligible results in reasonable time. StataCorp
>would want to protect users, and themselves, from results that were not
>defensible. That is the sort of thing they do a lot of.

It may also be protecting the user in that a Y with more than 50 
categories may have been specified by mistake, e.g. the user 
accidentally specified a continuous var rather than a categorical 
one.  I added limits to my own user-written programs precisely 
because a few people had made such mistakes.

Perhaps some sort of override option, e.g. maxcat(60), could be added 
in a future release of Stata.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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