Maximal Autocorrelation Functions in Functional Data Analysis

TL;DR

This paper introduces a novel factor rotation method for functional principal components analysis that enhances interpretability by emphasizing directions of decreasing smoothness, demonstrated through practical examples.

Contribution

It proposes a new rotation technique based on a generalized smoothing metric, improving interpretability of functional PCA components.

Findings

Rotation improves interpretability of principal components

Method is simple to implement

Demonstrated effectiveness on real data examples

Abstract

This paper proposes a new factor rotation for the context of functional principal components analysis. This rotation seeks to re-represent a functional subspace in terms of directions of decreasing smoothness as represented by a generalized smoothing metric. The rotation can be implemented simply and we show on two examples that this rotation can improve the interpretability of the leading components.