[R] multivariate lme?

[Constantinos Antoniou]

Dear all,

is it possible to estimate a multivariate multilevel model in R (I  
guess the term in R is mixed-effects model).  I can estimate  
univariate models using either lme{nlme} or lmer{lme4}, but when I  
use a multivariate response, I get the same output as if I was trying  
to estimate a univariate model.

[For a reproducible example, please download http://mit.edu/costas/ 
www/parav.txt and run:

 >Parav<-read.table("parav.txt",header=T)
 >library(nlme)
 >attach(Parav)
]


 >myResponse <- cbind(accidents,Killed)

I get the same thing as if I used only accidents as the response.

For example:

 >parav7.lme <- lme(accidents~alcontrols,random=~1+alcontrols| 
department/prefecture,data=Parav)

yields the same output as:

 >parav8.lme <- lme(myResponse~alcontrols,random=~1+alcontrols| 
department/prefecture,data=Parav)


Thank you,

Costas

PS. I am working on R-2.1.0a on a Mac 10.4.1.




--
Constantinos Antoniou, Ph.D.
Department of Transportation Planning and Engineering
National Technical University of Athens
5, Iroon Polytechniou str. GR-15773, Athens, Greece