Re: Mean centering for multicollinearity

Posted by Zack, Matthew M. (CDC/ONDIEH/NCCDPHP) on Aug 08, 2012; 12:04pm
URL: http://spssx-discussion.165.s1.nabble.com/Re-Mean-centering-for-multicollinearity-tp5714622.html

More than 20 years ago, David L. Belsley showed that mean centering does NOT remove multicollinearity
but only distorts the collinearity diagnostics used to detect it:

 Belsley DA.  Demeaning conditioning diagnostics through centering.  American Statistician 1984;
  38:73-93 (with comments and rejoinder).

 Belsley DA.  Conditioning diagnostics:  Collinearity and weak data in regression.  New York:
  John Wiley & Sons, Inc., 1991:175-204.

Matthew Zack

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD