[R-sig-ME] variance-covariance structure of random effects in glmer

[Douglas Bates]

On Mon, Jul 26, 2010 at 2:15 PM, Tanja Srebotnjak
<Tanja.Srebotnjak at ecologic-institute.us> wrote:
> Hello list members,

> I know I should study the documentation of lme4 but I think you might be
> able to answer my question in a blink: what is the default
> variance-covariance structure used in glmer in a logistic model with logit
> link and family binomial? And can it be changed?

I think you will need to refine your question a bit.  Are you asking
about the variance-covariance structure of the random effects, in
which case it depends on the random-effects terms in the model, or
about the variance-covariance of the conditional distribution of the
response, given the random effects and the values of the parameters.

Generally I would assume you are asking about the first possibility
but, as that doesn't depend on the family or the link, I'm not sure.

> Many thanks in advance!
> Tanja
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>