[Lme4-commits] Fitting different correlation structures under lme4

[jiangs at science.oregonstate.edu]

Hi everyone,

I was trying to fit my data using "lm" command from the R package  
"lme", however when I adopted the correlation structure ARMA with  
q=p=1, the process didn't converge. So I swithched to the new package  
"lme4".  I read through the whole illustration document as well as  
searched on internet for a long time, but I just couldn't figure out  
how to plug in a specific correlation structure such as ARMA, AR, or  
Exp spatial instead of an unstructured correlation matrix in the model  
fitting command "lmer". Does anybody have a suggestion?

Best,

Shuping