[R-sig-ME] error msg in nlme

[Douglas Bates]

On Feb 7, 2008 8:32 AM, Achaz von Hardenberg <fauna at pngp.it> wrote:
> hi all,
> Using the nlme package, I am doing some lme analyses on the effect of
> introduced fish on zooplankton biodiversity in alpine lakes.
> Everything going fine with all zooplankton taxa analyzed so far but
> one for which I get an error message which I do not know how to
> interpret:

>  > zoop.lme<-lme(Logcrpic ~ as.factor(year)+as.factor(month) + FAC1 +
> FAC2 + FAC3 + as.factor(fish),random=~1|
> Cod,data=zooplankton,na.action=na.omit)

> Error in solve.default(estimates[dimE[1] - (p:1), dimE[2] - (p:1),
> drop = FALSE]) :
>         system is computationally singular: reciprocal condition number =
> 2.21155e-16

The error is occurring in an attempt to invert a matrix - probably
when evaluating the standard errors of the estimates of the
fixed-effects coefficients.

I would suggest trying to fit the model with the lmer function in the
lme4 package instead.  Use

library(lme4)
zoop.lmer <- lmerlme(Logcrpic ~ as.factor(year)+as.factor(month) +
FAC1 +FAC2 + FAC3 + as.factor(fish) + (1|
Cod), data=zooplankton, verbose = TRUE)

> I would be thankful to anybody who could assist me in understanding
> what is going wrong with this particular variable...
>
> best regards,
> Achaz
>
> Dr. Achaz von Hardenberg
> ------------------------------------------------------------------------
> --------------------------------
> Centro Studi Fauna Alpina - Alpine Wildlife Research Centre
> Servizio Sanitario e della Ricerca Scientifica
> Parco Nazionale Gran Paradiso, Degioz, 11, 11010-Valsavarenche (Ao),
> Italy
>
>
>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>