[R] changes in the structure of mer objects?

Martin Maechler maechler at stat.math.ethz.ch
Mon Jan 15 14:00:12 CET 2007

Hi Joris,

I suspect you somehow load an older version lme4 or Matrix than
you think you are loading.
Or then you have an lmer() function or a class definition 
{from a saved workspace ????}
in your work space.

example(lmer) *must* run correctly for the 'lme4' package to get
onto CRAN at all,
hence it must be unique to your setup.

Maybe as a first step,
run R as "R --vanilla" ?

Martin Maechler, ETH Zurich

>>>>> "joris" == joris dewolf <joris.dewolf at cropdesign.com>
>>>>>     on Sun, 14 Jan 2007 19:15:43 +0100 writes:

    joris> Dear all,

    joris> I try to run the example of lmer and get the following error message.

    >> library(lme4)
    >> example(lmer)
    lmer> (fm1 <- lmer(Reaction ~ Days + (Days | Subject), sleepstudy))
    joris> [[1]]
    joris> Error in get(x, envir, mode, inherits) : variable "as.dpoMatrix" was not
    joris> found

    joris> This error message is similar to what I get with
    joris> other models. It looks like the mer class has a
    joris> slightly different structure. Anybody an idea how to
    joris> solve this?

    joris> I am using R 2.4.1 under linux and the latest releases of lme4 and Matrix

    joris> lme4_0.9975-10
    joris> Matrix_0.9975-8

    >> version
    joris> _
    joris> platform       x86_64-unknown-linux-gnu
    joris> arch           x86_64
    joris> os             linux-gnu
    joris> system         x86_64, linux-gnu
    joris> status
    joris> major          2
    joris> minor          4.1
    joris> year           2006
    joris> month          12
    joris> day            18
    joris> svn rev        40228
    joris> language       R
    joris> version.string R version 2.4.1 (2006-12-18)

More information about the R-help mailing list