[R] Multiv / pb with clustering methods?

[Charles Raux]

Le 31 Aug 00 à 12:26, Winfried Theis écrivait:

> you may try hclust in mva it does have plot.hclust as default plotting
> routine which does a dendrogram. 

plot.hclust with a result of hierclust produces a strange figure with 
an orphan vertical line (with my data) while with hclust it works 
well. 

> functions "diana" and "agnes" in the library cluster which do both hierarchicle
> clustering and have default plot routines which offer a dendrogram as well as
> another useful graphic called "banner" for details see Kaufman and Rousseeuw
> (1990) "Finding Groups in Data", Wiley, New York.

It seems impossible to use plot.agnes with a result of hierclust.
In fact my question about availability of dendrogram plotting *with 
hierclust* is motivated by a memory issue. As far as I understand 
hierclust is the only one which manages large data set (bign) (but I 
have not tried it yet with my "large" data sets (n>10,000). 

A more troubling issue is that I have checked the same data (18 obs, 
3 var) with the 3 functions and the same method (ward). This does 
not give the same results (height) and the same hierarchy:
> library(mva)
> library(multiv)
> library(cluster)
> h<-hclust(dist(col),method="ward")
> hc<-hierclust(data.matrix(col),method=1)
> ag<-agnes(col,method="ward")
> h$height
 [1] 0.005221242 0.009670544 0.010135010 0.017487202 0.019811790
 0.028691132 [7] 0.030891092 0.035334642 0.051179115 0.056848540
 0.057858657 0.073317828
[13] 0.152969145 0.228699576 0.269021808 0.659097143 0.756180840
> hc$height
 [1] 1.363068e-05 9.811893e-05 5.135922e-05 1.001958e+00 1.962535e-04
 [6] 3.071681e-04 4.115905e-04 1.301867e-03 8.769330e-04 1.061292e-03
[11] 1.673812e-03 1.859374e-03 9.444782e-03 2.431675e-02 2.924522e-02
[16] 7.656077e-02 9.686475e-02 
> ag$height
 [1] 0.005221242 0.046071504 0.009670544 0.029152501 0.220530047
 0.383204909 [7] 0.017487202 0.060143295 0.010135010 0.121114212
 0.028691132 0.423073789
[13] 0.019811790 0.035322842 0.051418163 0.183578860 0.057858657

Why?

---------------
Charles RAUX,
Laboratoire d'Economie des Transports
CNRS-Université Lumière Lyon 2-ENTPE
email : charles.raux at let.ish-lyon.cnrs.fr
http://www.ish-lyon.cnrs.fr/let
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._