English

Predictive density estimation under the Wasserstein loss

Statistics Theory 2021-09-01 v1 Statistics Theory

Abstract

We investigate predictive density estimation under the L2L^2 Wasserstein loss for location families and location-scale families. We show that plug-in densities form a complete class and that the Bayesian predictive density is given by the plug-in density with the posterior mean of the location and scale parameters. We provide Bayesian predictive densities that dominate the best equivariant one in normal models.

Keywords

Cite

@article{arxiv.1904.02880,
  title  = {Predictive density estimation under the Wasserstein loss},
  author = {Takeru Matsuda and William E. Strawderman},
  journal= {arXiv preprint arXiv:1904.02880},
  year   = {2021}
}
R2 v1 2026-06-23T08:30:01.874Z