(no subject)

Hi,
Yes, it does deal with mixture distributions, But in the sense I was
talking
about, random effects being nuisance parameters, one is in in effect
integrating them out
of the likelihood and hence it is a mixture, no? 
As I said, I am learning more than I am contributing to this discussion.

Nicholas
On Fri, 23 Jan 2009 15:58:56 -0800, "Albyn Jones" <jones at reed.edu> said:

On Fri, Jan 23, 2009 at 01:52:26PM -0600, Douglas Bates wrote:

Would Bruce Lindsay's work on the geometry of mixtures be applicable in
the mixed model setting?

I can't say because I am not familiar with that work.

I don't have it handy, but as I recall, it deals with mixture
distributions, not the mixed model.

albyn