[R] Redundancy in Variables (and canonical correlation)

Barker, Chris Barker at medtap.com
Fri Mar 21 15:19:45 CET 2003


 Base on the earlier question: there is a "redundancy" measure for
Canonical Correlation "Canonical Correlation Redundancy Analysis". You
may be able to adapt it to your situation. 
 
See the 
paper by 
Stewart David W., Love W. (1968), A General Canonical Correlation Index,
Psychological Bulletin, 70, 160-163.

There are dozens of references to it and "canonical correlation
redundancy "
on a Google Search (many are pointers to the online SAS help). 





            Chris Barker
   Director of Statistical Research
    MEDTAP International, Inc.
           Redwood City, Ca
 
        www.medtap.com
 
          650 632 4218

>



More information about the R-help mailing list