The Canonical Correlation Analysis

  • Antonio NavarraEmail author
  • Valeria Simoncini


The method based on the Singular Value Decomposition described in previous chapters was able to represent the largest amount of cross-covariance with fewest modes. The computed modes are designed so as to specifically explain most of the spatial variance. The spatial view point is a requirement that can be put on the modes, but it is hardly unique. Another commonly employed point of view arises if one considers the temporal variations of the modes. In this case one is interested in modes that generate maximally correlated time coefficients. This method has been developed in multi-variate analysis and it is known as “Canonical Correlation Analysis”


Weight Vector Canonical Correlation Analysis Height Field Time Coefficient Pacific North American Pattern 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Copyright information

© Springer Science+Business Media B.V. 2010

Authors and Affiliations

  1. 1.Ist. Nazionale di Geofisica e VulcanologiaBolognaItaly
  2. 2.Dip. to MatematicaUniversità di BolognaBolognaItaly

Personalised recommendations