Analyzing Highly Correlated Data

aultk.mmcri@mmc.org
Wed, 23 Aug 95 9:24:00 -0500

I am beginning to write a paper in which I propose using rotation of the axes of
a two dimensional histogram in order to "remove the correlation" from highly
correlated data. This seems to me to be a better way to analyze such data.

I think that what I am doing is probably similar to "principal components"
analysis, but it is less general and (hopefully) easier to understand (for clinical
types - you know ;))

Can anyone point me to references that may have dealt with this type of data
analysis? I probably have not explained it very well here, but would appreciate
any help I can get from the participants in this forum.

----------------------------------------------------------------
Kenneth A. Ault M.D.
Maine Medical Center Research Institute
125 John Roberts Road
South Portland ME 04106
Voice (207) 761-9090
Fax (207) 761-2130
E-Mail aultk.mmcri@office.mmc.org
----------------------------------------------------------------