[R-sig-ME] within_and_between_group_variance

[C. AMAL D. GLELE]

Thank you so much for your reply.
My question is not about a specific context;
about these terms, I've read somethings that were not clear for me ; so
that, I decide to request the help of the list to know more about them.
Thanks, again.
Best,



Le mar. 10 sept. 2019 à 16:01, Ben Bolker <bbolker using gmail.com> a écrit :

>   Can we have a little more context for the question please? Utility
> for what purpose?
>
>   GLMs (generalized linear models) don't have grouping structure at
> all, so the terms "within-group" and "between-group" don't really
> apply.
>
>   (G)LMMs summarize "among-group variance" as the variances of the
> random effects. "Within-group variance" is the residual variance for a
> linear mixed model (LMM, i.e. a GLMM with Gaussian response and
> identity link), harder to compute for non-LMM GLMMs (but papers e.g.
> by Nakagawa and Schielzeth define these components on the way to
> estimating R^2 or intra-class correlation values for GLMMs).
>
> On Tue, Sep 10, 2019 at 11:57 AM C. AMAL D. GLELE <altessedac2 using gmail.com>
> wrote:
> >
> > Hello, dear all.
> > Can somebody,  please, tell me:
> > among the outputs of a glmm and/or glm, what are:
> > 1) within group variance and it's utility
> > 2) between group variance and it's utility
> > In advance, many thanks.
> > Best regards,
> >
> >         [[alternative HTML version deleted]]
> >
> > _______________________________________________
> > R-sig-mixed-models using r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>

	[[alternative HTML version deleted]]