s-news
[Top] [All Lists]

Re: mixed multinomial model: factor coding

To: s-news@wubios.wustl.edu, hans.coppenolle@agr.kuleuven.ac.be
Subject: Re: mixed multinomial model: factor coding
From: Prof Brian Ripley <ripley@stats.ox.ac.uk>
Date: Wed, 21 Mar 2001 15:39:11 +0000 (GMT)
Reply-to: Prof Brian Ripley <ripley@stats.ox.ac.uk>
> Date: Wed, 21 Mar 2001 12:57:13 CET
> X-Priority: 3 (Normal)
> To: <s-news@wubios.wustl.edu>
> From: <hans.coppenolle@agr.kuleuven.ac.be> (Hans Coppenolle)
> Subject: [S] mixed multinomial model: factor coding
> X-Incognito-SN: 297
> X-Incognito-Version: 5.0.1.93
> 
> Hello,
> 
> I want to fit a mixed proportional odds model...
> My categorical response has 3 levels...
> I have 2 factorial factors (with 2 levels) as independent variables...
> My independent variables are random (subject-specific)...
> 
> How do i code my independent factors??
> Dummy coding (0-1) or (-1,1)???
> Can i treat my intercept as fixed??

What software are you using?  E.g. polr in library MASS does all the
coding for you.


-- 
Brian D. Ripley,                  ripley@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 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


<Prev in Thread] Current Thread [Next in Thread>