Hi, I was wondering what the best equivalent to SAS's FASTCLUS and PROC CLUSTER would be. I need to be able to test the significance of the clusters by comparing the probability of obtaining an equal or greater pseudo F to the Bonferroni-corrected level. I will also need to plot r squared against the number of clusters. Thanks so much, Taisa [[alternative HTML version deleted]]
At the R command prompt ?kmeans (for info on the R equivalent to FASTCLUS) ?hclust (for info on the R equivalent to CLUSTER) Install package clusterSim and look at function index.G1 for the Calinski-Harabasz pseudo F-statistic ---------------------------------------------- David L Carlson Associate Professor of Anthropology Texas A&M University College Station, TX 77843-4352> -----Original Message----- > From: r-help-bounces at r-project.org [mailto:r-help-bounces at r- > project.org] On Behalf Of Taisa Brown > Sent: Saturday, April 14, 2012 7:29 PM > To: r-help at r-project.org > Subject: [R] Cluster Analysis > > Hi, > > I was wondering what the best equivalent to SAS's FASTCLUS and PROC > CLUSTER would be. I need to be able to test the significance of the > clusters by comparing the probability of obtaining an equal or greater > pseudo F to the Bonferroni-corrected level. I will also need to plot r > squared against the number of clusters. > > Thanks so much, > > Taisa > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting- > guide.html > and provide commented, minimal, self-contained, reproducible code.
Hi, Taisa,
It depends on many paramfactors, e.g. nature of your data, volume of data set
etc.
The analog of SAS fastclus in R - kmeans (for practical example check slide #35
here:
http://www.slideshare.net/whitish/textmining-with-r)
Check also kmedoids (pam) and hclust.
Good luck,
-Alex
________________________________________
From: r-help-bounces at r-project.org [r-help-bounces at r-project.org] on
behalf of Taisa Brown [taisa.brown at unb.ca]
Sent: 15 April 2012 03:28
To: r-help at r-project.org
Subject: [R] Cluster Analysis
Hi,
I was wondering what the best equivalent to SAS's FASTCLUS and PROC CLUSTER
would be. I need to be able to test the significance of the clusters by
comparing the probability of obtaining an equal or greater pseudo F to the
Bonferroni-corrected level. I will also need to plot r squared against the
number of clusters.
Thanks so much,
Taisa
[[alternative HTML version deleted]]
______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.