Skip to content
Back to formatted view

Raw Message

Message-ID: <4238C5AC.7060505@adelaide.edu.au>
Date: 2005-03-16T23:47:56Z
From: David J. Netherway
Subject: principal component analysis
In-Reply-To: <1416.10.0.10.126.1110955951.squirrel@poisson.statisticon.se>

Hello,

I have used "prcomp" and the variances for the first 3 PC's are 2.65, 
1.97 and 0.38.

When I plot the principal component values for each data point I can see 
that the points lie in a plane as one might expect from the variances.

But this plane is diagonal through the 3D space of the first 3 PC's.  My 
question is: is it usual to follow a PC analysis with an eigen analysis 
to align this plane with with 2 principal directions, or is there a way  
of incorporating this into "prcomp".

Thanks, David