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
Locked
