DISCARDING VARIABLES in PRINCIPAL COMPONENT ANALYSIS : ALGORITHMS for ALL-SUBSETS COMPARISONS
The traditional approach to the interpretation of the results from a Principal Component Analysis implicitly discards variables that are weakly correlated with the most important and/or most interesting Principal Components. Some authors argue that this practice is potentially misleading and that it would be preferable to take a variable selection approach comparing variable subsets according to appropriate approximation criteria. In this paper, we propose algorithms for the comparison of all possible subsets according to some of the most important criteria proposed to date. The computational effort of the proposed algorithms is studied and it is shown that, given current computer technology, they are feasible for problems involving up to 30 variables. A software implementation is freely available on the internet.
|Date of creation:||Jan 2000|
|Date of revision:|
|Contact details of provider:|| Postal: |
Phone: +351 226 196 200
Fax: +351 226 196 291
Web page: http://www.feg.porto.ucp.pt/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:cap:wpaper:022000. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Ricardo Goncalves)
If references are entirely missing, you can add them using this form.