[R-sig-ME] Problem with glmer

[Ben Bolker]

On 13-12-03 09:21 AM, Hamada Elsayed Ali wrote:
> Dear All,
> 
> I am trying to run glmer models to test the effect of noncrop % and
> management effect on P/A data with 27000 records. I used the
> following code:
> 
> glmer100<-glmer(Present~Management+NonCrop100+(1|Plot.No), 
> family=binomial, data=Presence)
> 
> and it works fine, but when I tried to add the species name as
> follows:
> 
> glmer100<-glmer(Present~Management+NonCrop100+Short.Name+(1|Plot.No),
>  family=binomial, data=Presence)
> 
> It gave me the following warning:
> 
> Warning in rankMatrix(X) : rankMatrix(<large sparse Matrix>, method =
> 'tolNorm2') coerces to dense matrix. Probably should rather use
> method = 'qrLINPACK' !? Can any one help me,

   This is a harmless warning that can be ignored (a false positive from
the Matrix package); it could be eliminated by upgrading your version of
the Matrix package, after which you should at least reinstall (and
possibly update) your version of lme4.

  Ben Bolker