Skip to content
Prev 643 / 20628 Next

multivariate mixed model

It's not easy, but it can be done.  The approach is to define each
multivariate response as a cluster.  I did it in nlme for the
following article:

Robinson, A.P., 2004. Preserving correlation while modelling diameter
distributions.  Canadian Journal of Forest Research 34, 221

Daniel Hall (University of Georgia, I think) did it for non-linear
models, in a paper published in Biometrics (I think) about 5 years ago
...

Andrew
On Mon, Mar 10, 2008 at 07:58:35PM +0100, Irene Mantzouni wrote: