[R-sig-ME] grouped data or multiple responses

Douglas Bates bates at stat.wisc.edu
Mon Jan 26 20:02:38 CET 2009

On Mon, Jan 26, 2009 at 11:50 AM, Juan Pedro Steibel <steibelj at msu.edu> wrote:
> Hello everyone,
> Suppose I want to fit the same mixed model to a set of response variables
> and store the results in a single object.  Then retrieve the results one by
> one using an index or some other referencing method.

> The response may be stored either in multiple columns of a response matrix
> (no missing data, same design for all responses) or in a single vector, with
> an index variable indicating the response (allowing for different sampling
> schemes).

> Can anyone please help with some pointers on this?
> Thanks in advance,

Fit the first model then use refit with each subsequent data vector.  See


More information about the R-sig-mixed-models mailing list