Measuring association with Wasserstein distances

DOI10.3150/21-BEJ1438MaRDI QIDQ6359447arXivFDO

Authors Johannes Wiesel

Publication date 30 January 2021

Mathematics Subject Classification ID

Nonparametric inference (62Gxx) Multivariate analysis (62Hxx) Mathematical programming (90Cxx)

Abstract: Let $piinPi(mu,u)$ be a coupling between two probability measures $mu$ and $u$ on a Polish space. In this article we propose and study a class of nonparametric measures of association between $mu$ and $u$, which we call Wasserstein correlation coefficients. These coefficients are based on the Wasserstein distance between $u$ and the disintegration $p{i}_{x_1}$ of $pi$ with respect to the first coordinate. We also establish basic statistical properties of this new class of measures: we develop a statistical theory for strongly consistent estimators and determine their convergence rate in the case of compactly supported measures $mu$ and $u$. Throughout our analysis we make use of the so-called adapted/bicausal Wasserstein distance, in particular we rely on results established in [Backhoff, Bartl, Beiglb"ock, Wiesel. Estimating processes in adapted Wasserstein distance. 2020]. Our approach applies to probability laws on general Polish spaces.

This page was built for publication: Measuring association with Wasserstein distances

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6359447)