Dear list,
I am trying to fit some mixed models using packages lme4 and nlme.
I did the model selection using lmer but I suspect that I may have some
autocorrelation going on in my data so I would like to have a look using the
handy correlation structures available in nlme.
The problem is that I cannot translate my lmer model to lme:
mod1<- lmer(y~x + (1|a:b) + (1|b:c), data=mydata)
"a", "b" and "c" are factors with "c" nested in "b" and "b" nested in "a"
The best I can do with lme is:
mod2<- lme(y~x, random=list(a=~1, b=~1, c=~1), data=mydata)
which is the same as:
lmer(y~x + (1|a) + (1|a:b) + (1|a:b:c), data=mydata)
I am not at all interested in random effects (1|a) and (1|a:b:c) as they are
not significant. I just need two random intercepts as specified in mod1. How
can I translate mod1 into lme language?
Any help on this would be much appreciated.
Jonas
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
