[R] Is it possible with two random effects in lme()?

[Ben Bolker]

Bettina Lado Lindner <betti_la <at> hotmail.com> writes:


> Hello, 
> I want two random effect in R. One is BLOQUE nested in REP 
> and the other is TRAT. I don’t know how put TRAT
> independant of the other random effect.

[snip]

> I know that in lmer is easier but I use error correlation 
> so I need use lme.
> Bettina

  This (crossed random effects) is possible but not that easy: 
you should (1) study the
later chapters of Pinheiro and Bates 2000 and (2) re-post this
question to r-sig-mixed-models at r-project.org.

  good luck,
   Ben Bolker