Hello Murray,
thanks for the response. I would actually love to hear alternative
suggestions about the problem I am trying to solve. I just thought a
short question will be less of a burden on people's time and have a
higher chance of being answered.
basically the data sets I need to analyze contain 2000-1 objects.
each characterized by, depending on the data set, 9-20 attributes. all
integers greater than zero, typically the range is [0,1000] with numbers
5 particularly common. there is no apriori reason why these objects
should cluster into discrete groups. and in fact when the data is
explored graphically (xgobi) it doesn't show an obvious clustering
pattern. however, with 9-20 dimensions involved, it is probably easy to
miss subtle patterns. I have tried clustering the data using a number of
standard approaches including hclust,kmeans,fanny etc. but these methods
didn't seem to be able to generate convincingly distinct, homogeneous
clusters. of course given the type of the data involved Poisson mixtures
seem like the natural choice.
I have experimented a bit with snob using contrived data sets (where you
know which class objects really belong to) and it has been fairly
promising, except maybe for snob's tendency to break the known classes
into multiple subclasses.
I actually would like to try to code this in R. It would be very helpful
to me in fact if you can contribute any code/code fragments/examples
from your earlier work on this, either to the list or privately.
many thanks
Murad
[EMAIL PROTECTED] wrote:
The list could probably be more useful if you gave more details about your
data and the problem. I have written a bit of R code myself for fitting a
finite mixture of univariate Poissons by EM and found it very simple to
program in R. I suspect that your problem is multivariate, but that should
not present any difficulties.
The Snob program employs a fairly sophisticated model search strategy
based on the Minimum Message Length criterion. If you do not know much
about the solution that you are seeking it might be a good way to go. I
appreciate that Snob can be rather complex to set up and get going but I
think that you should be able to get quite a bit of help from the Monash
University people behind the program. They are usually quite keen to
encourage new users of Snob.
Murray Jorgensen
Hello,
I was wondering whether a Poisson mixture modeler/cluster analysis
package is available for R. I scanned CRAN packages and couldn't find
anything but I thought I'd ask. If not could anyone recommend a non-R
open source package. I have found 'snob' but this program seems a bit
hard to use in an automated, non interactive fashion.
regards,
Murad
--
Murad Nayal M.D. Ph.D.
Department of Biochemistry and Molecular Biophysics
College of Physicians and Surgeons of Columbia University
630 West 168th Street. New York, NY 10032
Tel: 212-305-6884 Fax: 212-305-6926
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
--
Murad Nayal M.D. Ph.D.
Department of Biochemistry and Molecular Biophysics
College of Physicians and Surgeons of Columbia University
630 West 168th Street. New York, NY 10032
Tel: 212-305-6884 Fax: 212-305-6926
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html