Canonical Correlation

From: Agustin Lobo (
Date: Tue Nov 03 1998 - 11:57:46 MET

One general question on Canonical Correlation:

Is it correct to use it in a predictive way?

That is:

We establish a correlation between a set of predictors
and a set of responses. Then, given a new observed response,
can we calculate the values of the predictors?

In other words, is it Canonical Correlation or
Canonical Regression?


Dr. Agustin Lobo
Instituto de Ciencias de la Tierra (CSIC)
Lluis Sole Sabaris s/n
08028 Barcelona SPAIN
tel 34 93409 5410
fax 34 93411 0012

This archive was generated by hypermail 2b30 : Sat Feb 10 2001 - 10:22:02 MET