Talk:Principal Components Analysis

I have started the PCA-wikipage as suggested by Markus. Apologies for the rough-looking first-draft. --Nikos 00:26, 7 March 2009 (UTC)

Some appearance editing--Nikos 14:18, 8 March 2009 (UTC)

Thanks for the idea.--Nikos 04:54, 16 March 2009 (UTC)

Added examples with SPOT from the Spearfish dataset. Confused about i.pca's performance: there seems to be no problem with the SPOT images, that is, the results of i.pca are almost identical with the results of R's prcomp() *with* centering. This is different than the example using MODIS data. Perhaps it depends on the range of the input data?!--Nikos 04:45, 31 March 2009 (UTC)

ToDo

• Need to update the "numbers" derived from i.pca (bug-fix 39800 gives correct numbers, identical to R's prcomp() function). --Nikos 05:26, 9 May 2010 (UTC)

• Complete review of the content, check terminology (especially for SVD, i.e. use "singular values, vectors" where and when ever required) --Nikos 20:24, 21 July 2010 (UTC)