[R] A repeated measures, linear mixed model (lme) WITHOUT random effects...

Mark Na mtb954 at gmail.com
Tue Aug 26 01:59:35 CEST 2008


Hello,

I am trying to fit a repeated measures linear mixed model (using lme) 
but I don't want to include any random effects. I'm having trouble (even 
after consulting Pinheiro & Bates 2000) figuring out how to specify the 
repeated measure without including it in the specification of a random 
effect.

My data consist of repeated "counts" in "plots" that I wish to model as 
a function of "habitat". This attempt:

          model<-lme(count~habitat-1,data=dataframe,method="ML")

doesn't consider the repeated nature of my counts (i.e., that there are 
multiple rows in my dataframe for each plot). I know how to 
include"plot" as a random effect, but I don't wish to do that, and I 
can't see how to include it without doing so.

I'll appreciate any help you can provide.

Thanks, Mark



More information about the R-help mailing list