[R-sig-ME] Estimation of random effects in generalized linear models

[Kotecha, Gopal]

Dear all,

I am trying to figure out how lme4 estimates individual random effect 
terms (the "b" terms in the following document 
https://cran.r-project.org/web/packages/lme4/vignettes/Theory.pdf).

My understanding is that the random effect standard deviation and fixed 
effects are first estimated, plugged into the likelihood and then this 
quantity is maximized (or some transformation to that effect).  Is this 
correct?

(More specifically I am figuring out how BLME does it, and BLME obtains 
estimates from lme4)

I have been through the "Computational Methods for Mixed Models" 
document 
(https://cran.r-project.org/web/packages/lme4/vignettes/Theory.pdf) but 
am not sure if it describes how estimation is carried out in the 
/generalized/ linear model case.

Thanks in advance for your time,

Gopal

	[[alternative HTML version deleted]]