Robert Dodier wrote:
> ...
> over all values of parameters. A particularly troublesome
> problem is choosing the number of components in the fitted
> mixture; typically there is a broad range of the number of
> components which fit more or less well. A scheme has been
> invented by Geoffrey Hinton and one of his students, called
> "splitting & merging expectation-maximization" (SMEM) in which
> changes of the number of components is accepted or rejected by
> a rule of the Metropolis type. The readers of the newsgroup
> may wish to take a look.
Beside split&merge techniques, there are also a large number of
criteria for choosing the "optimal" number of components,
proposed for example by McLachlan, by Raftery or by Celeux and
their respective colleagues. Depending on whether the goal is
density estimation or clustering, for example, the choice of
the criteria might be different.
Cyril.
---
Cyril Goutte [EMAIL PROTECTED]
INRIA Rhone-Alpes Tel: (+33) 4 76 61 55 13
Zirst - 655 avenue de l'Europe - Montbonnot Fax: (+33) 4 76 61 54 77
38334 Saint Ismier Cedex - France www.inrialpes.fr/is2
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
