PCA components are orthogonal by definition so no, that doesn't make
sense at all. Do yourself a favor and get a book on multivariate data
analysis. Two books come to mind: Obviously the one of Hair and
colleagues, called "multivariate data analysis" and easily found in a
university library (or on the internet).
The second one is "An R and S-Plus Companion to Multivariate Analysis"
by Everitt. This one you have less chance of finding in the library,
but you find it online for sale.
This is a nice introduction as well :
https://netfiles.uiuc.edu/miguez/www/Teaching/MultivariateRGGobi.pdf
Never use a chainsaw without reading the manual.
Cheers
Joris
On Tue, Jul 6, 2010 at 9:07 PM, Marino Taussig De Bodonia, Agnese
<agnese.marino09 at imperial.ac.uk> wrote:> Hello,
>
> I am currently analyzing responses to questionnaires about general
attitudes. I have performed a PCA on my data, and have retained two Principal
Components. Now I would like to use the scores of both the principal comonents
in a multiple regression. I would like to know if it makes sense to use the
scores of one principal component to explain the variance in the scores of
another principal component:
>
> lm(scores of principal component 1~scores of principal component 2+ age,
gender, etc..)
>
> Please could you let me know if this is statistically sound?
>
> Thank you in advance for you time,
>
> Agnese
> ______________________________________________
> 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.
>
--
Joris Meys
Statistical consultant
Ghent University
Faculty of Bioscience Engineering
Department of Applied mathematics, biometrics and process control
tel : +32 9 264 59 87
Joris.Meys at Ugent.be
-------------------------------
Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php