A Jacobian inequality for gradient maps on the sphere and its application to directional statistics

TL;DR

This paper establishes a Jacobian inequality for gradient maps on the sphere, demonstrating log-concavity of the Jacobian determinant with applications to defining new probability densities in directional statistics.

Contribution

It proves a novel Jacobian inequality for gradient maps on the sphere using the sphere's cross-curvature property, with implications for statistical density modeling.

Findings

Jacobian determinant is log-concave on the sphere

New family of probability densities defined via cost-convex functions

Likelihood functions exhibit log-concavity

Abstract

In the field of optimal transport theory, an optimal map is known to be a gradient map of a potential function satisfying cost-convexity. In this paper, the Jacobian determinant of a gradient map is shown to be log-concave with respect to a convex combination of the potential functions when the underlying manifold is the sphere and the cost function is the distance squared. The proof uses the non-negative cross-curvature property of the sphere recently established by Kim and McCann, and Figalli and Rifford. As an application to statistics, a new family of probability densities on the sphere is defined in terms of cost-convex functions. The log-concave property of the likelihood function follows from the inequality.