Dear R users, I m quite a novice in using R for factor analysis and I would need some help to choose the right function. I have a contingency table and I would like to perform a Correspondence analysis on this table, followed by a hirarchical clustering of my variables projected in on the first principal components. Here are my question : - what is the more appropriate function to do so ... I already tried by using the function 'corresp' in the MASS package - it seems to work ... but how is it possible to get all the information concerning the non-centered PCA used by corresp (eigen-values, inertia, scree plot, square cos, ....) In a second step I would like to use hirarchical clustering from the results of the correspondence analysis ... If "Table" is my contingency table (i.e., the number of individuals seen in each case) ... I tried to implment it as follow (for the 2 first components for instance, but I would be interesting at looking to the other components ...I did not manage to get the eigen values !) : A <- corresp(Table, nf = 2) biplot(A) hc <- hclust(dist(A$cscore), "ward") plot(hc) Is that ok ??? ... here it is exemplae using Ward method for clustering ... I would also be interested in using method of mutual neighbors to identify the clusters ... is it possible using hclust ? Any help would be really appreciated, Best regards ... Christophe *************************** Christophe Grova, PhD PostDoc - EEG department Montreal Neurological Institute, McGill University 3801 University Street, Montreal, Quebec, Canada, H3A 2B4 email : christophe.grova@mail.mcgill.ca tel : (514) 398 2184 fax : (514) 398 8106 web: http://idm.univ-rennes1.fr/users/grova *************************** [[alternative HTML version deleted]]
Martin Henry H. Stevens
2004-Jun-22 18:27 UTC
[R] Need for advise for Correspondence Analysis
You may want to look at the documentation for the vegan package. Hank Stevens On Jun 22, 2004, at 2:19 PM, christophe grova wrote:> > Dear R users, > > I m quite a novice in using R for factor analysis and I would need > some help to choose the right function. > I have a contingency table and I would like to perform a > Correspondence analysis on this table, followed by a hirarchical > clustering of my variables projected in on the first principal > components. > > Here are my question : > > - what is the more appropriate function to do so ... I already tried > by using the function 'corresp' in the MASS package > - it seems to work ... but how is it possible to get all the > information concerning the non-centered PCA used by corresp > (eigen-values, inertia, scree plot, square cos, ....) > > In a second step I would like to use hirarchical clustering from the > results of the correspondence analysis ... > > If "Table" is my contingency table (i.e., the number of individuals > seen in each case) ... I tried to implment it as follow (for the 2 > first components for instance, but I would be interesting at looking > to the other components ...I did not manage to get the eigen values !) > : > > A <- corresp(Table, nf = 2) > biplot(A) > hc <- hclust(dist(A$cscore), "ward") > plot(hc) > > Is that ok ??? ... here it is exemplae using Ward method for > clustering ... > I would also be interested in using method of mutual neighbors to > identify the clusters ... is it possible using hclust ? > > Any help would be really appreciated, > > Best regards ... > > Christophe > > *************************** > Christophe Grova, PhD > PostDoc - EEG department > Montreal Neurological Institute, McGill University > 3801 University Street, Montreal, Quebec, Canada, H3A 2B4 > email : christophe.grova at mail.mcgill.ca > tel : (514) 398 2184 > fax : (514) 398 8106 > web: http://idm.univ-rennes1.fr/users/grova > *************************** > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch 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 > >Dr. Martin Henry H. Stevens, Assistant Professor 338 Pearson Hall Botany Department Miami University Oxford, OH 45056 Office: (513) 529-4206 Lab: (513) 529-4262 FAX: (513) 529-4243 http://www.cas.muohio.edu/botany/bot/henry.html http://www.muohio.edu/ecology/ http://www.muohio.edu/botany/ "E Pluribus Unum"