[R] polr versus multinom

[Prof Brian Ripley]

Both polr and multinom are part of the support software for a book.  That
book covers your question.  Please read it, as you were asked to do.

On Thu, 28 Oct 2004, Schmidt.Matthias (FORST) wrote:

> I am searching for methods to compare regression models with an ordered
> categorical response variable (polr versus multinom).
> The pattern of predictions of both methods (using the same predictor
> variables) is quite different and the AIC is smaller for the multinom
> approach. I guess polr has more strict premises for the structure of the
> response variable, which methods can be used to test for these premises. If
> the AIC is smaller for the multinom approach is this approach more
> appropriate in every case?

> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

  `If the function is from a package accompanying a book, e.g., the MASS 
   package, consult the book before posting.'

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595