Theoretically and computationally convenient geometries on full-rank correlation matrices

TL;DR

This paper introduces new Riemannian geometries on full-rank correlation matrices that ensure unique logarithms and means, facilitating statistical analysis beyond the traditional SPD matrix framework.

Contribution

It generalizes the quotient-affine metric for correlation matrices and proposes three new Riemannian metrics with Hadamard structures, ensuring uniqueness of key operations.

Findings

New Riemannian metrics with Hadamard structures for correlation matrices

Closed-form expressions for Riemannian and group operations

Ensured uniqueness of logarithm and Fréchet mean in proposed geometries

Abstract

In contrast to SPD matrices, few tools exist to perform Riemannian statistics on the open elliptope of full-rank correlation matrices. The quotient-affine metric was recently built as the quotient of the affine-invariant metric by the congruence action of positive diagonal matrices. The space of SPD matrices had always been thought of as a Riemannian homogeneous space. In contrast, we view in this work SPD matrices as a Lie group and the affine-invariant metric as a left-invariant metric. This unexpected new viewpoint allows us to generalize the construction of the quotient-affine metric and to show that the main Riemannian operations can be computed numerically. However, the uniqueness of the Riemannian logarithm or the Fr{\'e}chet mean are not ensured, which is bad for computing on the elliptope. Hence, we define three new families of Riemannian metrics on full-rank correlation matrices which provide Hadamard structures, including two flat. Thus the Riemannian logarithm and the Fr{\'e}chet mean are unique. We also define a nilpotent group structure for which the affine logarithm and the group mean are unique. We provide the main Riemannian/group operations of these four structures in closed form.