[R-sig-ME] Missing information in "name" column of Random effects output

[Douglas Bates]

On 9/25/07, David Afshartous <dafshartous at med.miami.edu> wrote:

> Jorge,

> Awhile back I think I noticed a similar issue. I don't recall if I was able
> to do anything different w/ the lmer statement to fix it. In any event, my
> guess is that the first term corresponds to the intercept RE and the second
> to the slope RE.  Looking at the estimated RE's themselves supports this:

Jorge is correct that the name column is no longer being filled out
and David is correct that the order of variance components corresponds
to the order of the terms in the model matrix for the random effects
term.  Thanks to both of you for the report.

At some point I changed where the names are stored and I forgot to
update the VarCorr method accordingly.  I will release a new version
with this fixed.

> > ranef(fm1)
> An object of class ³ranef.lmer²
> [[1]]
>     (Intercept)        Days
> 308   2.2713451   9.1966961
> 309 -40.3825921  -8.6223099
> 310 -38.9403378  -5.4521894
> 330  23.6656443  -4.8100611
> 331  22.2391295  -3.0662861
> 332   9.0324969  -0.2709760
> 333  16.8277808  -0.2214775
> 334  -7.2256113   1.0733822
> 335  -0.3503176 -10.7492136
> 337  34.8784025   8.6302064
> 349 -25.1898574   1.1699674
> 350 -13.0500361   6.6107496
> 351   4.5697621  -3.0138529
> 352  20.8539011   3.5376139
> 369   3.2744289   0.8723798
> 370 -25.5865985   4.8179542
> 371   0.8049003  -0.9877807
> 372  12.3075594   1.2851975

> > fm1.junk = ranef(fm1)[[1]]
> > int.ranef  = fm1.junk[,1]
> > slope.ranef  = fm1.junk[,2]
> > var(int.ranef)
> [1] 465.8012
> > var(slope.ranef)
> [1] 29.75055

> On 9/25/07 2:42 PM, "Jorge González" <jorge.gonzalez at psy.kuleuven.be> wrote:
>
> > Dear all,
> >
> > My apologies if this has been discussed before. When estimating a model
> > with lmer, I can't see anymore the contents of the "name" column in the
> > Random effects part of the output. See the example below
> >
> > ###### begin example ###################
> >> sessionInfo()
> > R version 2.5.1 (2007-06-27)
> > i386-pc-mingw32
> >
> > locale:
> > LC_COLLATE=Dutch_Belgium.1252;LC_CTYPE=Dutch_Belgium.1252;LC_MONETARY=Dutch_Be
> > lgium.1252;LC_NUMERIC=C;LC_TIME=Dutch_Belgium.1252
> >
> > attached base packages:
> > [1] "stats"     "graphics"  "grDevices" "utils"     "datasets"
> > "methods"   "base"
> >
> > other attached packages:
> >         lme4       Matrix      lattice
> >  "0.99875-7" "0.999375-2"     "0.16-5"
> >> fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
> >> fm1
> > Linear mixed-effects model fit by REML
> > Formula: Reaction ~ Days + (Days | Subject)
> >    Data: sleepstudy
> >   AIC  BIC logLik MLdeviance REMLdeviance
> >  1754 1770 -871.8       1752         1744
> > Random effects:
> >  Groups   Name Variance  Std.Dev. Corr
> >  Subject             610.835  24.7151
> >                             35.056   5.9208  0.067
> >  Residual            655.066  25.5943
> > number of obs: 180, groups: Subject, 18
> >
> > Fixed effects:
> >                  Estimate   Std. Error t value
> > (Intercept)  251.405      6.820   36.86
> > Days            10.467      1.546    6.77
> >
> > Correlation of Fixed Effects:
> >      (Intr)
> > Days -0.137
> >>
> > #################### end example ##################
> > How can I know which variance correspond to the subject  random
> > intercept and which one to the subject slope Days? As far as I remember
> > this information was available in previous versions. Is this a bug?
> >
> > Thank you very much in advance
> >
> > Jorge
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>