[R-sig-ME] response correlated among groups

[Ben Bolker]

I think I would handle this as a stacked multivariate model:
http://www.rpubs.com/bbolker/3336 gives some examples.

On Wed, Nov 12, 2014 at 5:57 AM, Rafael Wüest <rafael.wueest at gmail.com>
wrote:

> Dear list members
>
> I would like to fit a model relating a gaussian response variable y to
> some continuous predictor variables (p1 and p2). The response is correlated
> among the classes of a factor (p3).
>
> Is it possible to set up a mixed model that accounts for the correlations
> among classes of p3 in the response?
>
> I would appreciate help on how to code such a model, if possible.
>
> Many thanks in advance,
> Rafael
>
> --
> Rafael Wüest
> rafael.wueest at gmail.com
> http://www.rowueest.net
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>

	[[alternative HTML version deleted]]