[R] How to personalize principal components analysis

Prof Brian Ripley ripley at stats.ox.ac.uk
Mon Sep 13 08:23:56 CEST 2004

On Mon, 13 Sep 2004, Simone Vantini wrote:

> Hallo!
> Does anybody know if it is possible to choose the function to be maximazed
> in the principal components analysis?Thanks

Principal components analysis is defined by a series of properties.  Two 

1) Maximize the variance of the projection onto a kD subspace

2) Minimize the sum of squares of the distances from points to their 
projections onto a kD subspace.

So the criteria are what defines PCA.  Please consult the references in 
e.g. ?princomp.

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

More information about the R-help mailing list