On universally consistent and fully distribution-free rank tests of vector independence
From MaRDI portal
Publication:6344421
DOI10.1214/21-AOS2151arXiv2007.02186WikidataQ114060478 ScholiaQ114060478MaRDI QIDQ6344421
Mathias Drton, Marc Hallin, Hongjian Shi, Unnamed Author
Publication date: 4 July 2020
Abstract: Rank correlations have found many innovative applications in the last decade. In particular, suitable rank correlations have been used for consistent tests of independence between pairs of random variables. Using ranks is especially appealing for continuous data as tests become distribution-free. However, the traditional concept of ranks relies on ordering data and is, thus, tied to univariate observations. As a result, it has long remained unclear how one may construct distribution-free yet consistent tests of independence between random vectors. This is the problem addressed in this paper, in which we lay out a general framework for designing dependence measures that give tests of multivariate independence that are not only consistent and distribution-free but which we also prove to be statistically efficient. Our framework leverages the recently introduced concept of center-outward ranks and signs, a multivariate generalization of traditional ranks, and adopts a common standard form for dependence measures that encompasses many popular examples. In a unified study, we derive a general asymptotic representation of center-outward rank-based test statistics under independence, extending to the multivariate setting the classical H'{a}jek asymptotic representation results. This representation permits direct calculation of limiting null distributions and facilitates a local power analysis that provides strong support for the center-outward approach by establishing, for the first time, the nontrivial power of center-outward rank-based tests over root- neighborhoods within the class of quadratic mean differentiable alternatives.
Nonparametric hypothesis testing (62G10) Asymptotic properties of nonparametric inference (62G20) Hypothesis testing in multivariate analysis (62H15) Measures of association (correlation, canonical correlation, etc.) (62H20)
This page was built for publication: On universally consistent and fully distribution-free rank tests of vector independence
