s-news
[Top] [All Lists]

Re: probit/logit model with random effect

To: "Essioux, Laurent {PRG~Basel}" <laurent.essioux@roche.com>
Subject: Re: probit/logit model with random effect
From: Prof Brian Ripley <ripley@stats.ox.ac.uk>
Date: Mon, 27 Sep 2004 14:02:55 +0100 (BST)
Cc: Karin <twentestat@yahoo.com>, <s-news@wubios.wustl.edu>
In-reply-to: <C47BDB1BC5F5D9489E3457513B71A17601391020@rbamsem1.emea.roche.com>
On Mon, 27 Sep 2004, Essioux, Laurent {PRG~Basel} wrote:

> You can download a .pdf file from the nlme package in R where everything
> is explained in details regarding logistic mixed models

But that is unlikely to be what is meant by `logit model with random 
effect', to quote the title.

Both senses are discussed with examples in MASS4, which you, Laurent, 
could read to find out the differences.

> Laurent 
> 
> -----Original Message-----
> From: s-news-owner@lists.biostat.wustl.edu
> [mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Prof Brian
> Ripley
> Sent: Monday, September 27, 2004 2:41 PM
> To: Karin
> Cc: s-news@wubios.wustl.edu
> Subject: Re: [S] probit/logit model with random effect
> 
> 
> On Mon, 27 Sep 2004, Karin wrote:
> 
> > Is is possible to analyse a logit (or probit) model in S-plus with a 
> > random effect included?
> 
> Yes.
> 
> Depending which meaning of `logit model' you have in mind see either
> nlme 
> or glmmPQL (in library section MASS).
> 
> 

-- 
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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


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