[R-sig-ME] lme4 missing from repositories?
Federico Calboli
f.calboli at imperial.ac.uk
Wed Oct 20 18:30:14 CEST 2010
On 20 Oct 2010, at 16:29, Douglas Bates wrote:
> On Wed, Oct 20, 2010 at 10:18 AM, huang min <minhuangr at gmail.com> wrote:
>> Hi, Prof. Bates,
>>
>> I tried some simple codes from Prof. Ripley.
>>
>> library(lme4)
>> y <- (1:20)*pi; x <- (1:20)^2;group <- gl(2,10)
>> M2. <- lmer (y ~ 1 + x + (1 + x | group))
>> M2 <- lmer (y ~ x + ( x | group))
>> identical(fixef(M2), fixef(M2.))
>
> This is exactly what is so frustrating. By the time the model
> matrices have been constructed M2. and M2 are exactly the same model
> so how they end up with different answers is somewhat mysterious and
> something that we can't reproduce on other systems.
FWIW:
R version 2.12.0 (2010-10-15)
Copyright (C) 2010 The R Foundation for Statistical Computing
ISBN 3-900051-07-0
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
....
[R.app GUI 1.35 (5632) x86_64-apple-darwin9.8.0]
library(lme4)
test = rep(0, 10000)
for (i in 1:10000){
y <- (1:20)*pi; x <- (1:20)^2;group <- gl(2,10)
M2. <- lmer (y ~ 1 + x + (1 + x | group))
M2 <- lmer (y ~ x + ( x | group))
test[i] = identical(fixef(M2), fixef(M2.))
}
sum(test)
[1] 10000
sum(test == F)
[1] 0
I just fail to see why there is any need to use R32 and not R64, but that's just me.
F
--
Federico C. F. Calboli
Department of Epidemiology and Biostatistics
Imperial College, St. Mary's Campus
Norfolk Place, London W2 1PG
Tel +44 (0)20 75941602 Fax +44 (0)20 75943193
f.calboli [.a.t] imperial.ac.uk
f.calboli [.a.t] gmail.com
More information about the R-sig-mixed-models
mailing list