Skip to content

problem in the implementation of the function "lmer" of the "lme4" package - Important

3 messages · cristiana vieira, Ben Bolker, Martin Maechler

Original
3 days later
Ben Bolker
# 
cristiana vieira <cristiana1710vieira at ...> writes:
[snip]
(Sorry about the mangling of the example format below.)
  I very strongly suspect that this is due to a mismatch with
the version of the Matrix package.  Version 1.0-5 is available
on CRAN; could you please install it (either via update.packages(),
or via install.packages(), or via the appropriate menu item
if you are running in a GUI -- they should all be equivalent)
and report back if the problem persists?
  
   (The listed dependency of the package is only
Matrix >=0.9996875-1 ; we should probably update this, although
we are also working hard to replace the current version of lme4 ...)

  Ben Bolker
Martin Maechler
# 
> cristiana vieira <cristiana1710vieira at ...> writes:
    >> 
    >> 

    > [snip]
 
    >> When i ran the example given in the help of the software
    >> R and I obtained the same error: > library(lme4)Loading
    >> required package: MatrixLoading required package: lattice
    >> Attaching package: ?Matrix? The following object(s)
    >> are masked from ?package:base?: det Attaching
    >> package: ?lme4? The following object(s) are masked
    >> from ?package:stats?: AIC, BIC > ?lmerstarting
    >> httpd help server ... done > (fm1 <- lmer(Reaction ~ Days
    >> + (Days|Subject), sleepstudy))Error in as(ff,
    >> "sparseMatrix") : unused argument(s) ("sparseMatrix") >
    >> sessionInfo()R version 2.14.2 (2012-02-29)Platform:
    >> i386-pc-mingw32/i386 (32-bit) locale:[1]
    >> LC_COLLATE=Portuguese_Portugal.1252
    >> LC_CTYPE=Portuguese_Portugal.1252 [3]
    >> LC_MONETARY=Portuguese_Portugal.1252 LC_NUMERIC=C [5]
    >> LC_TIME=Portuguese_Portugal.1252 attached base
    >> packages:[1] stats graphics grDevices utils datasets
    >> methods base other attached packages:[1] lme4_0.999375-42
    >> Matrix_1.0-4 lattice_0.20-0 loaded via a namespace (and
    >> not attached):[1] grid_2.14.2 nlme_3.1-103 stats4_2.14.2
    >> tools_2.14.2


    >   (Sorry about the mangling of the example format below.)
    > I very strongly suspect that this is due to a mismatch
    > with the version of the Matrix package.  Version 1.0-5 is
    > available on CRAN; could you please install it (either via
    > update.packages(), or via install.packages(), or via the
    > appropriate menu item if you are running in a GUI -- they
    > should all be equivalent) and report back if the problem
    > persists?

Hmm,  R-2.14.2  *comes* with Matrix 1.0-4 
*included* (as it is a recommended package).

And many other people have been using the same combination
without any such fundamental problems

Somehow you must really have done something bad to your R (2.14.2)
installation, or to your lme4 installation.  Maybe you should
reinstall R,  or at least re-install lme4 -- for *that* version of R.
I don't think you should reinstall Matrix.

  
    >    (The listed dependency of the package is only Matrix
    > >=0.9996875-1 ; we should probably update this,

we have updated it for lme4.0
however, lme4 (the current one on CRAN) has worked with many
versions of Matrix in the past, and has not depended
specifically on any recent version of Matrix.

    > although we are also working hard to replace the current version of
    > lme4 ...)
[indeed!]