[R-sig-ME] Fwd: failure notice

[Felix Bach]

-
Hello,

I am new to this group and hope that my question is not too simple. however, I checked previous messages and could not find an answer to my question. 
Iwould like to obtian the credibility interval for the intraclass correlation for a random efffects model. For example

(fm1 <- lmer(Reaction ~ Days + (1|Subject), sleepstudy))

Linear mixed model fit by REML 
Formula: Reaction ~ Days + (1 | Subject) 
   Data: sleepstudy 
Random effects:
 Groups   Name        Variance Std.Dev.
 Subject  (Intercept) 1378.18  37.124  
 Residual              960.46  30.991  

--> ICC= 1378/(1378+960)

set.seed(101); samp0 <- mcmcsamp(fm1, n = 1000)


str(samp0)

Formal class 'merMCMC' [package "lme4"] with 9 slots
  ..@ Gp      : int [1:2] 0 18
  ..@ ST      : num [1, 1:1000] 1.198 1.104 0.941 0.805 0.874 ...
  ..@ call    : language lmer(formula = Reaction ~ Days + (1 | Subject), data = sleepstudy)
  ..@ deviance: num [1:1000] 1794 1794 1794 1795 1798 ...
  ..@ dims    : Named int [1:18] 1 180 2 18 1 1 0 1 2 5 ...
  .. ..- attr(*, "names")= chr [1:18] "nt" "n" "p" "q" ...
  ..@ fixef   : num [1:2, 1:1000] 251.4 10.5 256.8 10.1 263.1 ...
  .. ..- attr(*, "dimnames")=List of 2
  .. .. ..$ : chr [1:2] "(Intercept)" "Days"
  .. .. ..$ : NULL
  ..@ nc      : int 1
  ..@ ranef   : num[1:18, 0 ] 
  ..@ sigma   : num [1, 1:1000] 31 30 31.9 30.9 31.1 ...

I can get the residual variance (@sigma) but not the subject variance.

Any suggestions would be highly appreciated. Many thanks
Felix
-- 
GMX DSL SOMMER-SPECIAL: Surf & Phone Flat 16.000 für nur 19,99 ¿/mtl.!*

-- 
GMX DSL SOMMER-SPECIAL: Surf & Phone Flat 16.000 für nur 19,99 ¿/mtl.!*