[R] p values in lme4 package

Andrew Robinson A.Robinson at ms.unimelb.edu.au
Sat Mar 24 19:09:34 CET 2007


Dear Cristina,

On Sat, Mar 24, 2007 at 01:51:34PM +0100, Cristina Gomes wrote:
> Dear R-users,
> I was wondering if anybody knows if it's possible to obtain a p value 
> for the full model of a GLMM with the lme4 package. 

I do not believe that it is possible to do so.

> I was told that I 
> should check whether the full model including all the predictor 
> variables is significant before doing stepwise regression or further 
> analysis, but I can't figure out how to do this. I also wanted to know 
> if there's a way of obtaining residuals or predicted values for the same 
> analysis.

There is, but it is not automatic.  You have to write your own code to
compute the residuals and predicted values for each model, using the
estimates of the fixed and random effects.  lmer is still under
extensive development by its author, and the "helper" functions to
extract or compute residuals and predicted values have not yet been
written.

Cheers

Andrew

-- 
Andrew Robinson  
Department of Mathematics and Statistics            Tel: +61-3-8344-9763
University of Melbourne, VIC 3010 Australia         Fax: +61-3-8344-4599
http://www.ms.unimelb.edu.au/~andrewpr
http://blogs.mbs.edu/fishing-in-the-bay/



More information about the R-help mailing list