[R] sampling weights for lme

[alex pegucci]

Dear All,

I have a complex survey data with observations having differing
probabilities of selection into the sample. I would like to run a
linear mixed effects model and also to use weighting that takes this
into consideration. As far as I know, however, the weighting option for
the lme command from nlme package (or glmmPQL from MASS, for that
matter) is related to heteroscedasticity. Is there a sampling weights
option for these packages that I am not aware of? 

Regards,

Alex