[R-sig-ME] ranef() on Mac OS X

[Douglas Bates]

Notice that you have different versions of the lme4 package on Ubuntu
and on the Mac.  For some reason the lme4 package is failing a test on
Mac OS X only and we haven't been able to diagnose why the -35 version
fails there.  Also, we haven't been able to reproduce this on other
operating systems and we can't decide why it should fail on Mac OS X
unless it is an obscure memory issue.  As neither Martin nor I use Mac
OS X we aren't able to debug the issue.  Other folks have looked at it
but all they can say is that it is failing, no idea why.  We even
tried to bypass the test on the Mac but that doesn't seem to have
worked either.

I'm open to suggestions from the list on what can be done about lme4
and Mac OS X.  If you look at the test results for the package on
CRAN, e.g.

http://cran.us.r-project.org/web/checks/check_results_lme4.html

you will see that Mac OS X is the only system on which there is a
problem and the test on which it fails doesn't make sense for it to
fail.  The only difference between the two models being compared is
that they were generated from slightly different formulas but the
models are identical once the formula is parsed.

On Tue, Oct 19, 2010 at 11:31 AM, Christopher Desjardins
<desja004 at umn.edu> wrote:
> I am unable to use the ranef() function on a lmer function on a Mac. I get
> the following error:
>
>> fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
>> ranef(fm1)
> Error in UseMethod("ranef") :
>  no applicable method for 'ranef' applied to an object of class "mer"
>> sessionInfo()
> R version 2.11.1 (2010-05-31)
> i386-apple-darwin9.8.0
>
> locale:
> [1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] plyr_1.2.1         nlme_3.1-96        lme4_0.999375-34
> Matrix_0.999375-39 lattice_0.18-8     foreign_0.8-40
>
> loaded via a namespace (and not attached):
> [1] grid_2.11.1   stats4_2.11.1 tools_2.11.1
>
> This works as expected on Ubuntu 10.10 64 bit (see below). I built the lme4
> package from source on the Mac and I know that lme4 has had some issues on
> Mac for a while (i.e. it hasn't been available as a binary for some time). I
> just wanted to post this here to see if this was either a new bug, something
> already documented, or something that I've overlooked. Googling 'ranef lme4
> Mac' brought me to this thread
> https://stat.ethz.ch/pipermail/r-sig-mixed-models/2010q3/004091.html ... but
> the author was able to use the ranef() function.
>
> Thanks,
> Chris
>
>
> #################
> ### From Linux ###
> #################
>> fm1 <- lmer(Reaction ~ Days + (1|Subject), sleepstudy)
>> ranef(fm1)
> $Subject
>    (Intercept)
> 308   40.787403
> 309  -77.856603
> 310  -63.114282
> 330    4.406841
> 331   10.217114
> 332    8.221982
> 333   16.501988
> 334   -2.997253
> 335  -45.286227
> 337   72.189222
> 349  -21.198168
> 350   14.112641
> 351   -7.862933
> 352   36.381719
> 369    7.037018
> 370   -6.363279
> 371   -3.294571
> 372   18.117387
>> sessionInfo()
> R version 2.11.1 (2010-05-31)
> x86_64-pc-linux-gnu
>
> locale:
>  [1] LC_CTYPE=en_US.utf8       LC_NUMERIC=C
>  [3] LC_TIME=en_US.utf8        LC_COLLATE=en_US.utf8
>  [5] LC_MONETARY=C             LC_MESSAGES=en_US.utf8
>  [7] LC_PAPER=en_US.utf8       LC_NAME=C
>  [9] LC_ADDRESS=C              LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] lme4_0.999375-35   Matrix_0.999375-40 lattice_0.18-8
>
> loaded via a namespace (and not attached):
> [1] grid_2.11.1   nlme_3.1-96   stats4_2.11.1 tcltk_2.11.1  tools_2.11.1
>
>
>
>
> --
> Christopher David Desjardins
> Ph.D. student, Quantitative Methods in Education
> M.S. student, Statistics
> University of Minnesota
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>