[R] scores from multinomial logistic regression

[Prof Brian Ripley]

What do you mean by the scores?

What multinom does is to fit probabilities (which you can extract by 
fitted()): the response is a discrete probability distribution.
There is an underlying linear predictor but

(a) it is K-dimensional and
(b) there is a degree of ambiguity, usually resolved by setting the 
predictor for one category to zero (but not in this code).

That linear predictor is only generated in the underlying C code.

On Fri, 7 May 2004, Jacqueline Hall wrote:

> Dear all,
>  
> I'm interested in extracting the score from multinomial logistic regression
> models fit using multinom, to assess the stregth of assocation of the
> parameter with the response (akin to the score from clogit/cox regression).
> currently I'm using R 1.8.1.
> Is there a function that will extract the score from a multinom object or
> how i can get back to it? or from using glm?
> I investigated the documention for Design but those functions seem to apply
> to binary logistic.

-- 
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