[R-sig-ME] glmer

[PLUQUET Thibault]

Hi,

I'm a french student in Statistics. I have a question on generalized linear mixed models.For the estimation, we have to do the penalized least squares but I don't understand why this method ? I know why we use the penalized least squares for lieanr mixed models . But , I don't understand why there is the matrix W in this method and how to get the discrepancy function.

Thank you for your help, 


Sincerely 


PLUQUET Thibault