Nonlinear Canonical Correlation Analysis

The purpose of nonlinear canonical correlation analysis is to determine how similar two or more sets of variables are to one another. As in linear canonical correlation analysis, the aim is to account for as much of the variance in the relationships among the sets as possible in a low-dimensional space. Unlike linear canonical correlation analysis, however, nonlinear canonical correlation analysis does not assume an interval level of measurement or assume that the relationships are linear. Another important difference is that nonlinear canonical correlation analysis establishes the similarity between the sets by simultaneously comparing linear combinations of the variables in each set to an unknown set—the object scores.

Next