Using the Sinkhorn divergence in permutation tests for the multivariate two-sample problem
Statistics Theory
2022-09-30 v1 Methodology
Statistics Theory
Abstract
In order to adapt the Wasserstein distance to the large sample multivariate non-parametric two-sample problem, making its application computationally feasible, permutation tests based on the Sinkhorn divergence between probability vectors associated to data dependent partitions are considered. Different ways of implementing these tests are evaluated and the asymptotic distribution of the underlying statistic is established in some cases. The statistics proposed are compared, in simulated examples, with the test of Schilling's, one of the best non-parametric tests available in the literature.
Cite
@article{arxiv.2209.14455,
title = {Using the Sinkhorn divergence in permutation tests for the multivariate two-sample problem},
author = {E. del Barrio and J. S. Osorio and A. J. Quiroz},
journal= {arXiv preprint arXiv:2209.14455},
year = {2022}
}
Comments
33 pages, 14 figures