Re: Plotting a principal component analysis with contours

[Gerrit J. Kiers]

From: "Johan Kullstam" <address@hidden>

> for canonical variate analysis, you almost have to go with SVD.  tw
> anderson's eigenvalue derivation for CVA is imho incomprehensible.

I indeed consulted his 'multivariate statistical analysis'. I share your
conclusion. Thanks for the comfort! But of course my judgement is based on a
lack of education in this field and has little to do with Andersons
accomplishments.... ;-)

> anyhow, once you do an SVD you can look at the problem as multiple
> one- dimensional gaussians.  solve the 1D cases to get spheres of
> confidence intervals.  use the SVD mapping to get elipses in the
> original vector space.

Thanks for this help, I'll try this and will publish the files here once I'm
successful.

Gerrit






-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------