Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Model-based Clustering


From   "Daniel Stegmueller" <d.stegmueller@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Model-based Clustering
Date   Tue, 28 Aug 2007 12:38:42 +0200

Lina,

Stas is right - Fraley and Raftery's approach is implemented in R in
the 'mclust' package.
They also have a nice piece in JASA 97 (p. 611-31) explaining the
general procedure...

Best,
Daniel



2007/8/26, Stas Kolenikov <skolenik@gmail.com>:
> There were some commands to do normal mixtures (which you could try to
> -findit-), but Fraley & Raftery's approach is deeply Bayesian, which
> does not mesh up terribly well with Stata's concepts. I would guess
> they had their implementation in Splus, as there were no R around back
> in 1990s, but they may have ported in to R since then.
>
> On 8/24/07, Lina Jonsson <lina.jonsson@vti.se> wrote:
> > Hello.
> >
> > I know that there exist a function called mclust that works with S-PLUS for Model-based Cluster Analysis according to an article by Fraley and Raftery from 1998. Sense I´m no expert when it comes to clustering, it´s hard for me to be more specific. But in the article model-based cluster analysis is described as "In model-based cluster analysis data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geomteric properties are obtained through Gaussian compnents with different parameterixations and cross-cluster constraints..." I hope this gives you a clue to what I`m after. I appreciate all the help and suggestions I can get.
> >
> > Best Regards
> >
> > Lina Jonsson
> >
> >
> >
> >
> >
> > > -----Ursprungligt meddelande-----
> > > Från: owner-statalist@hsphsun2.harvard.edu
> > > [mailto:owner-statalist@hsphsun2.harvard.edu] För Marcello Pagano
> > > Skickat: den 24 augusti 2007 14:25
> > > Till: statalist@hsphsun2.harvard.edu
> > > Ämne: Re: st: Model-based Clustering
> > >
> > > Can you be a little more specific please, Lina?
> > > Clustering can have about three different meanings in a
> > > statistical context, and that's just off the top of my head,
> > > that are quite different.
> > >
> > > m.p.
> > >
> > > Lina Jonsson wrote:
> > > > Hello.
> > > >
> > > > I would like to do model-based clustering so I wonder if
> > > there are any
> > > > commands for doing this i Stata.
> > > >
> > > >
> > > > Best Regards
> > > >
> > > > Lina Jonsson
> > > >
> > > > *
> > > > *   For searches and help try:
> > > > *   http://www.stata.com/support/faqs/res/findit.html
> > > > *   http://www.stata.com/support/statalist/faq
> > > > *   http://www.ats.ucla.edu/stat/stata/
> > > >
> > > *
> > > *   For searches and help try:
> > > *   http://www.stata.com/support/faqs/res/findit.html
> > > *   http://www.stata.com/support/statalist/faq
> > > *   http://www.ats.ucla.edu/stat/stata/
> > >
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/support/faqs/res/findit.html
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> >
>
>
> --
> Stas Kolenikov, also found at http://stas.kolenikov.name
>
> Small print: Please do not reply to my Gmail address as I don't check
> it regularly.
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>


-- 
Friend to Groucho Marx:  "Life is difficult!"
Marx to Friend:  "Compared to what?"


Daniel Stegmüller
Böckstr. 11
68159 Mannheim
0621/9783639

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index