[R] standardized random effects with ranef.lme()

Dirk Enzmann dirk.enzmann at uni-hamburg.de
Mon Jul 24 22:04:57 CEST 2006

Using ranef() (package nlme, version 3.1-75) with an 'lme' object I can 
obtain random effects for intercept and slope of a certain level (say: 
1) - this corresponds to (say level 1) "residuals" in MLWin. Maybe I'm 
mistaken here, but the results are identical.

However, if I try to get the standardized random effects adding the 
paramter "standard=T" to the specification of ranef(), the results 
differ considerably from the results of MLWin (although MLWin defines 
"standardized" in the same way as "divided by its estimated (diagnostic) 
standard error").

Why do the results differ although the estimates (random effects and 
thus their variances) are almost identical? I noticed that lme() does 
not compute the standard errors of the variances of the random effects - 
for several reasons, but if this is true, how does ranef() calculate the 
standardized random effects (the help says: '"standardized" (i.e. 
divided by the corresponding estimated standard error)').

Is there a way to obtain similar results as in MLWin (or: should I 
prefer the results of ranef() for certain reasons)?


R version: 2.3.1 Patched (2006-06-21 r38367)

Dr. Dirk Enzmann
Institute of Criminal Sciences
Dept. of Criminology
Edmund-Siemers-Allee 1
D-20146 Hamburg

phone: +49-(0)40-42838.7498 (office)
        +49-(0)40-42838.4591 (Billon)
fax:   +49-(0)40-42838.2344
email: dirk.enzmann at uni-hamburg.de

More information about the R-help mailing list