[R] Principal components vs. raw variables

[Manuel Ramon]

Hello to everyone,

I am starting to work on classification procedures. I usualy do a principal
component analysis (PCA) as a previous step in order to reduce variables and
after I apply a cluster procedure. My question is if it will be better to
use raw variables instead of use principal components obtained from these
variables since the original variables keep all the variability.

Now i am thinking to use a variable group analysis (VGA) and a correlation
analysis together in order to identify which of my original variables could
explain differences on my data better, and after apply a cluster analysis on
selected variables. 

What do you think about it? What would be better: work with PCA or with raw
variables.

Thanks in advance.

Manuel



-----
Manuel Ramón Fernández
Group of Reproductive Biology (GBR)
University of Castilla-La Mancha (Spain)
mramon at jccm.es
-- 
View this message in context: http://www.nabble.com/Principal-components-vs.-raw-variables-tp22824280p22824280.html
Sent from the R help mailing list archive at Nabble.com.