Canonical analysis is hard to explain because it involves the difficult concepts of eigenvalue and eigenvector. In this note, I describe an approach that I have found to be effective in overcoming these difficulties which uses the new concept of an eigenangle to explain eigenanalysis.
No takes yet. Share an insight, caveat, or question.
Donald G. Watts (1995) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: