[R] PCA with spearman and kendall correlations

[David L Carlson]

Spearman would be easier since you just convert the data to ranks and use
the Pearson correlation:

> set.seed(42)
> x <- data.frame(matrix(sample(1:9, 20, replace=TRUE), 10, 2))
> x
   X1 X2
1   9  5
2   9  7
3   3  9
4   8  3
5   6  5
6   5  9
7   7  9
8   2  2
9   6  5
10  7  6
> cor(x)
           X1         X2
X1 1.00000000 0.01897427
X2 0.01897427 1.00000000
> cor(x, method="spearman")
            X1          X2
X1  1.00000000 -0.03135181
X2 -0.03135181  1.00000000
> cor(sapply(x, rank))
            X1          X2
X1  1.00000000 -0.03135181
X2 -0.03135181  1.00000000

----------------------------------------------
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 BOURGADE Eric
> Sent: Thursday, February 28, 2013 3:50 AM
> To: r-help at r-project.org
> Subject: [R] PCA with spearman and kendall correlations
> 
> Hello,
> 
> I would like to do a PCA with dudi.pca or PCA, but also with the use of
> Spearman or Kendall correlations
> Is it possible ?
> Otherwise, how can I do, according to you ?
> 
> Thanking you in advance
> 
> Eric Bourgade
> RTE
> France
> 
> 
> 
> 
> 
> 	[[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.