[R-sig-ME] RE : with lmer : Erreur dans names(argNew)[1] <- names(formals(new))[[1]]

[Douglas Bates]

The problem is subtle and traces back to your calling the new data
frame "new" thereby overriding the function new in the methods
package.   The right hand side of that assignment in the function
Matrix::fac2sparse should be changed to
names(formals(methods::new))[1] and we will make the change in the
next release of the Matrix package.  Until then you can run the script
if you change the name new to newDF

2010/8/4 DUYME Florent <F.DUYME at arvalisinstitutduvegetal.fr>:
> Here are data, script and sessionInfo()
> Florent
>
> R version 2.10.0 (2009-10-26)
> i386-pc-mingw32
>
> locale:
> [1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252    LC_MONETARY=French_France.1252 LC_NUMERIC=C
> [5] LC_TIME=French_France.1252
>
> attached base packages:
> [1] grDevices datasets  splines   graphics  stats     tcltk     utils     methods   base
>
> other attached packages:
>  [1] lme4_0.999375-33   Matrix_0.999375-41 lattice_0.18-3     car_2.0-2          nnet_7.3-1         MASS_7.3-4         svSocket_0.9-48    TinnR_1.0.3
>  [9] R2HTML_2.0.0       Hmisc_3.8-0        survival_2.35-7
>
> loaded via a namespace (and not attached):
> [1] cluster_1.13.1 grid_2.10.0    nlme_3.1-96    svMisc_0.9-57  tools_2.10.0
>
>
>
> -----Message d'origine-----
> De : dmbates at gmail.com [mailto:dmbates at gmail.com] De la part de Douglas Bates
> Envoyé : mardi 3 août 2010 17:52
> À : DUYME Florent
> Cc : r-sig-mixed-models at r-project.org
> Objet : Re: [R-sig-ME] with lmer : Erreur dans names(argNew)[1] <- names(formals(new))[[1]]
>
> Without a reproducible example it is difficult to determine why the
> problem may be manifesting itself.  I would note that you are using
> the car package and there may be an interaction between having car
> loaded and some of the methods in lme4.  However, without a
> reproducible example and some information on exactly what versions of
> which packages are attached (i.e. the output from sessionInfo()) this
> is just speculation.
>
>
> On Tue, Jul 20, 2010 at 3:18 AM, DUYME Florent
> <F.DUYME at arvalisinstitutduvegetal.fr> wrote:
>> Hi
>>
>>
>>
>> While running lmer after lm, I get  this error message :
>>
>>
>>
>> Erreur dans names(argNew)[1] <- names(formals(new))[[1]] :
>>
>>  l'argument de remplacement est de longueur nulle
>>
>>
>>
>> I don't understand why it appears. I re-start R, and then lmer works.
>>
>>
>>
>> I suppose the previous work has an influence on lmer; why ?
>>
>>
>>
>>
>>
>> My previous work:
>>
>> op<-options(contrasts=c("contr.sum", "contr.treatment"))
>>
>> resu1<-lm(Y ~ fact1 + bloc_c*bloc_l, data=DF)
>>
>> Anova(resu1, type="III")  # library car
>>
>> summary(resu1)
>>
>> options(op)
>>
>>
>>
>>
>>
>> sessionInfo()
>>
>> R version 2.10.0 (2009-10-26)
>>
>> i386-pc-mingw32
>>
>>
>>
>> locale:
>>
>> [1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252    LC_MONETARY=French_France.1252 LC_NUMERIC=C
>>
>> [5] LC_TIME=French_France.1252
>>
>>
>>
>> attached base packages:
>>
>> [1] grDevices datasets  splines   graphics  stats     tcltk     utils     methods   base
>>
>>
>>
>> other attached packages:
>>
>>  [1] lme4_0.999375-33   Matrix_0.999375-41 car_1.2-16         multcomp_1.1-7     mvtnorm_0.9-9      lattice_0.18-3     gdata_2.8.0        RODBC_1.3-1
>>
>>  [9] svSocket_0.9-48    TinnR_1.0.3        R2HTML_2.0.0       Hmisc_3.8-0        survival_2.35-7
>>
>>
>>
>> loaded via a namespace (and not attached):
>>
>> [1] cluster_1.13.1 grid_2.10.0    gtools_2.6.2   nlme_3.1-96    svMisc_0.9-57  tools_2.10.0
>>
>>
>>
>>
>>
>>
>>
>> thanks for your advices
>>
>>
>>
>> florent
>>
>>
>>        [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>