[R-sig-ME] lme4 broken ?

lancelot renaud.lancelot at cirad.fr
Fri Apr 13 12:26:00 CEST 2012


Dear Martin,

I just installed the patched version of R, and then re-installed from 
source the CRAN version of lme4. It now works fine. The same for the 
R-Forge version of lme4. Here is the session info for the latter case:

 > sessionInfo()
R version 2.15.0 Patched (2012-04-09 r58947)
Platform: i386-pc-mingw32/i386 (32-bit)

locale:
[1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252
[3] LC_MONETARY=French_France.1252 LC_NUMERIC=C
[5] LC_TIME=French_France.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] lme4_0.999902344-0 Matrix_1.0-7       lattice_0.20-6

loaded via a namespace (and not attached):
[1] grid_2.15.0    MASS_7.3-17    minqa_1.2.0    nlme_3.1-103 
splines_2.15.0
[6] tools_2.15.0

Thank you very much for your help.

Best regards,

Renaud

Le 13/04/2012 12:00, Martin Maechler a écrit :
> As Matrix comes with R,  could you try to just use the version
> that comes with R 2.15.0, and *then* install lme4 ?
> Or in any case, install lm4 after Matrix, as lme4 depends on
> Matrix?
>
> I agree that this is not "useful" behavior...
> Martin
>
>>>>>> lancelot<renaud.lancelot at cirad.fr>
>>>>>>      on Fri, 13 Apr 2012 11:49:20 +0200 writes:
>
>      >  Dear Martin and all,
>      >  Thank you very much for your time. However:
>
>      >>  remove.packages("lme4")
>      >  Elimination de(s) package(s) de ‘C:/R/RLIBS’
>      >  (car ‘lib’ n'est pas spécifié)
>      >>  install.packages("lme4") Installation d(es) package(s) dans
>      >  ‘C:/R/RLIBS’ (car ‘lib’ n'est pas spécifié) essai de l'URL
>      >  'http://cran.r-project.org/bin/windows/contrib/2.15/lme4_0.999375-42.zip'
>      >  Content type 'application/zip' length 1300772 bytes (1.2 Mb) URL
>      >  ouverte downloaded 1.2 Mb
>
>      >  le package ‘lme4’ a été décompressé et les sommes MD5 ont été vérifiées
>      >  avec succés
>
>      >  Les packages binaires téléchargés sont dans
>      >  C:\Temp\RtmpsnSVf7\downloaded_packages
>      >>  update.packages(checkBuilt = TRUE) Matrix : Version 1.0-5 installed in
>      >  C:/R/RLIBS built under R 2.15.0 Version 1.0-7 available at
>      >  http://R-Forge.R-project.org Update (y/N/c)?  y essai de l'URL
>      >  'http://R-Forge.R-project.org/bin/windows/contrib/2.15/Matrix_1.0-7.zip'
>      >  Content type 'application/zip' length 3252028 bytes (3.1 Mb) URL
>      >  ouverte downloaded 3.1 Mb
>
>      >  le package ‘Matrix’ a été décompressé et les sommes MD5 ont été
>      >  vérifiées avec succés
>
>      >  Les packages binaires téléchargés sont dans
>      >  C:\Temp\RtmpCGB7op\downloaded_packages
>      >>  library(lme4)
>      >  Le chargement a nécessité le package : lattice
>      >  Le chargement a nécessité le package : Matrix
>      >>  example(lmer)
>
>      lmer>  ## linear mixed models - reference values from older code
>      lmer>  (fm1<- lmer(Reaction ~ Days + (Days|Subject), sleepstudy))
>      >  Erreur dans diag(val) :
>      >  7 arguments passed to .Internal(identical) which requires 6
>
>
>      >  Renaud
>
>
>      >  Le 13/04/2012 09:19, Martin Maechler a écrit :
>      >>>>>>>  "l" == lancelot<renaud.lancelot at cirad.fr>
>      >>>>>>>  on Fri, 13 Apr 2012 08:55:24 +0200 writes:
>      >>
>      >>  >   I re-installed the previous version of Matrix (Matrix_1.0-5.tar.gz
>      >>  >   available on CRAN) from source, and it now works (see below).
>      >>
>      >>  >   I confirm there is a problem - at least on Windows - between lme4
>      >>  >   and R-Forge present versions) and the present version of
>      >>  >   Matrix 1.0-6.
>      >>
>      >>  Hmm, I very much doubt this, at it least in this generality.
>      >>  Other people have seen similar problems (not involving lme4 at
>      >>  all), on R-help or R-devel, and the solution has been to
>      >>  re-install packages (or R!) only.
>      >>
>      >>  The mismatch of number of arguments of .Internal(identical)
>      >>  is really from using packages that were *built* with one version
>      >>  of R, in a different version of *R*.
>      >>
>      >>  So, in principle
>      >>
>      >>  update.packages(checkBuilt = TRUE)
>      >>
>      >>  should get you there.
>      >>
>      >>  Regards,
>      >>  Martin Maechler
>      >>
>      >>
>      >>  >   All the best,
>      >>
>      >>  >   Renaud
>      >>
>      >>  >>   sessionInfo()
>      >>  >   R version 2.15.0 (2012-03-30)
>      >>  >   Platform: i386-pc-mingw32/i386 (32-bit)
>      >>
>      >>  >   locale:
>      >>  >   [1] LC_COLLATE=French_France.1252  LC_CTYPE=French_France.1252
>      >>  >   LC_MONETARY=French_France.1252 LC_NUMERIC=C
>      >>  >   LC_TIME=French_France.1252
>      >>
>      >>  >   attached base packages: [1] stats graphics grDevices utils datasets
>      >>  >   methods base
>      >>
>      >>  >   other attached packages:
>      >>  >   [1] lme4_0.999902344-0 Matrix_1.0-5       lattice_0.20-6
>      >>
>      >>  >   loaded via a namespace (and not attached):
>      >>  >   [1] grid_2.15.0    MASS_7.3-17    minqa_1.2.0    nlme_3.1-103
>      >>  >   splines_2.15.0 tools_2.15.0
>      >>
>      >>
>      >>  >>   library(lme4)
>      >>  >   Le chargement a nécessité le package : lattice
>      >>  >   Le chargement a nécessité le package : Matrix
>      >>  >>   example(lmer)
>      >>
>      lmer>  ## linear mixed models - reference values from older code
>      lmer>  (fm1<- lmer(Reaction ~ Days + (Days|Subject), sleepstudy))
>      >>  >   Linear mixed model fit by REML ['lmerMod']
>      >>  >   Formula: Reaction ~ Days + (Days | Subject)
>      >>  >   Data: sleepstudy
>      >>
>      >>  >   REML criterion at convergence: 1743.628
>      >>
>      >>  >   Random effects:
>      >>  >   Groups   Name        Variance Std.Dev. Corr
>      >>  >   Subject  (Intercept) 612.10   24.741
>      >>  >   Days         35.07    5.922   0.066
>      >>  >   Residual             654.94   25.592
>      >>  >   Number of obs: 180, groups: Subject, 18
>      >>
>      >>  >   Fixed effects:
>      >>  >   Estimate Std. Error t value
>      >>  >   (Intercept)  251.405      6.825   36.84
>      >>  >   Days          10.467      1.546    6.77
>      >>
>      >>  >   Correlation of Fixed Effects:
>      >>  >   (Intr)
>      >>  >   Days -0.138
>      >>
>      lmer>  (fm2<- lmer(Reaction ~ Days + (1|Subject) + (0+Days|Subject),
>      >>  >   sleepstudy))
>      >>  >   Linear mixed model fit by REML ['lmerMod']
>      >>  >   Formula: Reaction ~ Days + (1 | Subject) + (0 + Days | Subject)
>      >>  >   Data: sleepstudy
>      >>
>      >>  >   REML criterion at convergence: 1743.669
>      >>
>      >>  >   Random effects:
>      >>  >   Groups   Name        Variance Std.Dev.
>      >>  >   Subject  (Intercept) 627.58   25.052
>      >>  >   Subject  Days         35.86    5.988
>      >>  >   Residual             653.58   25.565
>      >>  >   Number of obs: 180, groups: Subject, 18
>      >>
>      >>  >   Fixed effects:
>      >>  >   Estimate Std. Error t value
>      >>  >   (Intercept)  251.405      6.885   36.51
>      >>  >   Days          10.467      1.560    6.71
>      >>
>      >>  >   Correlation of Fixed Effects:
>      >>  >   (Intr)
>      >>  >   Days -0.184
>      >>
>      lmer>  anova(fm1, fm2)
>      >>  >   Data: sleepstudy
>      >>  >   Models:
>      >>  >   fm2: Reaction ~ Days + (1 | Subject) + (0 + Days | Subject)
>      >>  >   fm1: Reaction ~ Days + (Days | Subject)
>      >>  >   Df    AIC    BIC  logLik deviance  Chisq Chi Df Pr(>Chisq)
>      >>  >   fm2  5 1762.0 1778.0 -876.00   1752.0
>      >>  >   fm1  6 1763.9 1783.1 -875.97   1751.9 0.0639      1     0.8004
>      >>
>      >>
>      >>
>      >>  >   Le 12/04/2012 20:11, Joehanes, Roby (NIH/NHLBI) [F] a écrit :
>      >>  >>   Hi Renaud,
>      >>  >>
>      >>  >>   It works okay in my computer (Linux). Maybe you want to update all
>      >>  >>   underlying libraries?
>      >>  >>
>      >>  >>   Hope this helps, Roby
>      >>  >>
>      >>  >>   (fm1<- lmer(Reaction ~ Days + (Days|Subject), sleepstudy))
>      >>  >>
>      >>  >>   _______________________________________________
>      >>  >>   R-sig-mixed-models at r-project.org mailing list
>      >>  >>   https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>      >>  >>
>      >>
>      >>  >   --
>      >>  >   Renaud Lancelot
>      >>  >   EDENext Project, coordinator: http://www.edenext.eu/
>      >>
>      >>  >   CIRAD, UMR15, Campus International de Baillarguet TA A-DIR / B
>      >>  >   F34398 Montpellier
>      >>
>      >>  >   Tel.  +33 4 67 59 37 17  -  Fax  +33 4 67 59 37 98
>      >>  >   Secr. +33 4 67 59 37 37  - Cell. +33 6 77 52 08 69
>      >>
>      >>  >   _______________________________________________
>      >>  >   R-sig-mixed-models at r-project.org mailing list
>      >>  >   https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>      >>
>
>      >  --
>      >  Renaud Lancelot
>      >  EDENext Project, coordinator: http://www.edenext.eu/
>
>      >  CIRAD, UMR15, Campus International de Baillarguet TA A-DIR / B
>      >  F34398 Montpellier
>
>      >  Tel.  +33 4 67 59 37 17  -  Fax  +33 4 67 59 37 98
>      >  Secr. +33 4 67 59 37 37  - Cell. +33 6 77 52 08 69
>

-- 
Renaud Lancelot
EDENext Project, coordinator: http://www.edenext.eu/

CIRAD, UMR15, Campus International de Baillarguet TA A-DIR / B
F34398 Montpellier

Tel.  +33 4 67 59 37 17  -  Fax  +33 4 67 59 37 98
Secr. +33 4 67 59 37 37  - Cell. +33 6 77 52 08 69



More information about the R-sig-mixed-models mailing list