[R] Re: Biplot function of PCA

Fboucays fboucays at wanadoo.fr
Mon Sep 9 09:15:29 CEST 2002


		    [was sent a wrong R-help address; manually resent by MM]
Hello
I'am using the 'biplot' and 'biplot.pincomp' functions of the 'mva'
package for my studies.

The biplot represents both the observations and the variables of a
matrix of multivariate data on the same plot.

The observations are represented by their numbers (the line of the data
matrix), but I would need to change it.

I would like to have the possibility, to represent on a biplot, the
observations, by their name, or , by the group or the cluster they
belong.

For instance:
tab is a data.frame (4 variables x 52 obsevations)

pc.tab_princomp(tab,cor=TRUE)

biplot.princomp(pc.tab)

km_kmeans(tab,3,100)

cl_cutree(hclust(dist(tab)),k=3)

# I would like to obtain:

biplot.princomp(pc.tab, nomi=cl)

#or

biplot.princomp(pc.tab,nomi=km$cluster)

Like that I could see how the clusters are spread on the PCA.

So, my question is,

Shall I write myself this new function ?

Perhaps someone already done-it? But how could I know.

Thank you taking your time for my problem.

Best regards

Francois BOUCAYS
*************************
Assistant Hospitalier Universitaire
Faculté de Chirurgie Dentaire
Sciences Anatomiques (Sous-Section 58.03)
3, ch. des Maraîchers - 31062 Toulouse Cedex
----------------------------
Laboratoire de Chirurgie Cephalo-Rachidienne Assistée par Ordinateur
Directeur: Pr F. Boutault
----------------------------
tel: 05.62.17.29.29 - fax: 05.61.25.47.19
port: 06.13.24.44.45 - emel: fboucays at wanadoo.fr
***************************************
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list