Re: [R] clustering: Multivariate t mixtures

From: Nicholas Lewin-Koh <nikko_at_hailmail.net>
Date: Fri 09 Sep 2005 - 07:13:05 EST

Hi,
Actually that was my plan was to implement a new flexmix class. Thanks for the pointer to the jss paper, that will be helpful.

Nicholas
On Thu, 8 Sep 2005 23:07:13 +0200, "Achim Zeileis" <Achim.Zeileis@wu-wien.ac.at> said:
> On Thu, 08 Sep 2005 15:38:55 -0500 Nicholas Lewin-Koh wrote:
>
> > Hi,
> > Before I write code to do it does anyone know of code for fitting
> > mixtures of multivariate-t distributions.
> > I can't use McLachan's EMMIX code because the license is "For non
> > commercial use only".
> > I checked, mclust and flexmix but both only do Gaussian.
>
> The Gaussian case is available in a pre-packaged function FLXmclust(),
> but the flexmix framework is not limited to that case. There is a paper
> which appeared in the Journal of Statistical Software
> (http://www.jstatsoft.org/) that explains how to write new M-steps for
> flexmix. It is also contained in the package as
> vignette("flexmix-intro")
>
> Best,
> Z
>
> > Thanks
> > Nicholas
> >
> > ______________________________________________
> > R-help@stat.math.ethz.ch mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide!
> > http://www.R-project.org/posting-guide.html
> >



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Fri Sep 09 07:25:45 2005

This archive was generated by hypermail 2.1.8 : Sun 23 Oct 2005 - 16:38:33 EST