[R] Can I use "lme" to deal with grouping data when I only get one data point per group?

From: Ruixiao Lu <tinypenguin_at_gmail.com>
Date: Thu 15 Sep 2005 - 04:43:17 EST


Hi there,
I have a question for using "lme". Say, I have 6 data points and they belong to six groups (one group factor). So there is no replicates for each group and I cannot separate the with-in group variation from the between group variation.
But when I try to use "lme" to deal with it, it gave the answers for both with-in group variation and the between group variation! The statement is as below:


fac=as.factor(c(1:6))

y=data

data.y=data.frame(y,fac)

y.g=groupedData(y~1|fac)

fit.y=lme(y~1,random=~1|fac)



An example is:
y=c(-0.3465181, -0.2019839, -0.7610653, -0.1992943, -0.1663348, 0.2811794) then the lme gave me the variance components: 0.09865809 (intercept , between-group variance) 0.01387379 (residual, with-in group variance)  So, my question is, from theory we cannot get separate estimates of with-in group variation and the between group variation, then what dose the output of the "lme" mean?

Thanks!
 Ruixiao

        [[alternative HTML version deleted]]



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed Sep 21 03:11:12 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:40:20 EST