Canonical correlation analysis
Canonical correlation analysis (CCA) is one of the most important tools of multivariate statistical analysis. Its extension to the functional context is not trivial, and in many ways illustrates the differences between multivariate and functional data. One of the most influential contributions has been made by Leurgans et al. (1993) who showed that smoothing is necessary in order to define the functional canonical correlations meaningfully.
Unable to display preview. Download preview PDF.