Bug in weights in lmer

Hello,

in case this if of any interest, I obtained apparently the same results as 
Nick using:
sessionInfo()
R version 2.6.2 (2008-02-08)
i386-pc-mingw32

locale:
LC_COLLATE=English_United Kingdom.1252;LC_CTYPE=English_United 
Kingdom.1252;LC_MONETARY=English_United 
Kingdom.1252;LC_NUMERIC=C;LC_TIME=English_United Kingdom.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] lme4_0.999375-13  Matrix_0.999375-9 lattice_0.17-6

loaded via a namespace (and not attached):
[1] grid_2.6.2  tools_2.6.2

### Results:
w<-rep(1,nrow(sleepstudy))
(fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy) )
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
   Data: sleepstudy
  AIC  BIC logLik deviance REMLdev
 1756 1775 -871.8     1752    1744
Random effects:
 Groups   Name        Variance Std.Dev. Corr
 Subject  (Intercept) 612.095  24.7405
          Days         35.071   5.9221  0.065
 Residual             654.944  25.5919
Number of obs: 180, groups: Subject, 18

Fixed effects:
            Estimate Std. Error t value
(Intercept)  251.405      6.825   36.84
Days          10.467      1.546    6.77

Correlation of Fixed Effects:
     (Intr)
Days -0.138

(fm2 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy, weights =w))
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
   Data: sleepstudy
  AIC  BIC logLik deviance REMLdev
 1756 1775 -871.8     1752    1744
Random effects:
 Groups   Name        Variance Std.Dev. Corr
 Subject  (Intercept) 6640265  2576.87
          Days         380465   616.82  0.065
 Residual             7105115  2665.54
Number of obs: 180, groups: Subject, 18

Fixed effects:
            Estimate Std. Error t value
(Intercept)    975.9      710.8   1.373
Days           173.3      161.0   1.076

Correlation of Fixed Effects:
     (Intr)
Days -0.138

(fm3 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy, weights 
=w/sum(w)) )
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
   Data: sleepstudy
   AIC BIC logLik deviance REMLdev
 820.9 840 -404.4    817.3   808.9
Random effects:
 Groups   Name        Variance  Std.Dev. Corr
 Subject  (Intercept) 169992420 13038.1
          Days          9740118  3120.9  0.065
 Residual               1010516  1005.2
Number of obs: 180, groups: Subject, 18

Fixed effects:
            Estimate Std. Error t value
(Intercept)    72.74    3596.51 0.02022
Days           12.92     814.62 0.01586

Correlation of Fixed Effects:
     (Intr)
Days -0.138

Cheers,

Luca

---------------------------
Luca B?rger, PhD
Postdoctoral Research Fellow
Department of Integrative Biology
University of Guelph
Guelph, Ontario, Canada N1G 2W1
phone: +1 519 824 4120 ext. 54554
fax:     +1 519 767 1656


----- Original Message ----- 
From: "Nick Isaac" <njbisaac at googlemail.com>
To: "Doran, Harold" <HDoran at air.org>
Cc: <R-sig-mixed-models at r-project.org>
Sent: Wednesday, April 23, 2008 12:00 PM
Subject: Re: [R-sig-ME] Bug in weights in lmer
Thanks Harold - very curious.
See my results below. This sounds like a versioning issue, in which
case it would be good to get as many folk as possible to replicate it.

cheers, Nick

w<-rep(1,nrow(sleepstudy))
(fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy) )
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
  Data: sleepstudy
 AIC  BIC logLik deviance REMLdev
1756 1775 -871.8     1752    1744
Random effects:
Groups   Name        Variance Std.Dev. Corr
Subject  (Intercept) 612.095  24.7405
         Days         35.071   5.9221  0.065
Residual             654.944  25.5919
Number of obs: 180, groups: Subject, 18

Fixed effects:
           Estimate Std. Error t value
(Intercept)  251.405      6.825   36.84
Days          10.467      1.546    6.77

Correlation of Fixed Effects:
    (Intr)
Days -0.138
 (fm2 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy, weights =
+ w) )
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
  Data: sleepstudy
 AIC  BIC logLik deviance REMLdev
1756 1775 -871.8     1752    1744
Random effects:
Groups   Name        Variance Std.Dev. Corr
Subject  (Intercept) 6640265  2576.87
         Days         380465   616.82  0.065
Residual             7105115  2665.54
Number of obs: 180, groups: Subject, 18

Fixed effects:
           Estimate Std. Error t value
(Intercept)    975.9      710.8   1.373
Days           173.3      161.0   1.076

Correlation of Fixed Effects:
    (Intr)
Days -0.138
(fm3 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy, weights =
+ w/sum(w)) )
Linear mixed model fit by REML
Formula: Reaction ~ Days + (Days | Subject)
  Data: sleepstudy
  AIC BIC logLik deviance REMLdev
820.9 840 -404.4    817.3   808.9
Random effects:
Groups   Name        Variance  Std.Dev. Corr
Subject  (Intercept) 169988830 13038.0
         Days          9739898  3120.9  0.065
Residual               1010497  1005.2
Number of obs: 180, groups: Subject, 18

Fixed effects:
           Estimate Std. Error t value
(Intercept)    72.74    3596.47 0.02022
Days           12.92     814.61 0.01586

Correlation of Fixed Effects:
    (Intr)
Days -0.138

_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

Bug in weights in lmer

Thread (4 messages)