English

Minimax bounds for estimating multivariate Gaussian location mixtures

Statistics Theory 2021-05-20 v2 Statistics Theory

Abstract

We prove minimax bounds for estimating Gaussian location mixtures on Rd\mathbb{R}^d under the squared L2L^2 and the squared Hellinger loss functions. Under the squared L2L^2 loss, we prove that the minimax rate is upper and lower bounded by a constant multiple of n1(logn)d/2n^{-1}(\log n)^{d/2}. Under the squared Hellinger loss, we consider two subclasses based on the behavior of the tails of the mixing measure. When the mixing measure has a sub-Gaussian tail, the minimax rate under the squared Hellinger loss is bounded from below by (logn)d/n(\log n)^{d}/n. On the other hand, when the mixing measure is only assumed to have a bounded pthp^{\text{th}} moment for a fixed p>0p > 0, the minimax rate under the squared Hellinger loss is bounded from below by np/(p+d)(logn)3d/2n^{-p/(p+d)}(\log n)^{-3d/2}. These rates are minimax optimal up to logarithmic factors.

Keywords

Cite

@article{arxiv.2012.00444,
  title  = {Minimax bounds for estimating multivariate Gaussian location mixtures},
  author = {Arlene K. H. Kim and Adityanand Guntuboyina},
  journal= {arXiv preprint arXiv:2012.00444},
  year   = {2021}
}
R2 v1 2026-06-23T20:38:14.066Z