Open Access
2022 Minimax bounds for estimating multivariate Gaussian location mixtures
Arlene K. H. Kim, Adityanand Guntuboyina
Author Affiliations +
Electron. J. Statist. 16(1): 1461-1484 (2022). DOI: 10.1214/21-EJS1975

Abstract

We prove minimax bounds for estimating Gaussian location mixtures on Rd under the squared L2 and the squared Hellinger loss functions. Under the squared L2 loss, we prove that the minimax optimal rate is upper and lower bounded by a constant multiple of n1(logn)d2. 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)dn, which implies that the optimal minimax rate is between (logn)dn and the upper bound (logn)d+1n obtained by [11]. On the other hand, when the mixing measure is only assumed to have a bounded pth moment for a fixed p>0, the minimax rate under the squared Hellinger loss is bounded from below by np(p+d)(logn)3d2. This rate is minimax optimal up to logarithmic factors.

Funding Statement

The first author was supported by NRF-2020R1F1A1A01069632. The second author was supported by NSF CAREER Grant DMS-16-54589.

Citation

Download Citation

Arlene K. H. Kim. Adityanand Guntuboyina. "Minimax bounds for estimating multivariate Gaussian location mixtures." Electron. J. Statist. 16 (1) 1461 - 1484, 2022. https://doi.org/10.1214/21-EJS1975

Information

Received: 1 May 2021; Published: 2022
First available in Project Euclid: 2 March 2022

MathSciNet: MR4387848
zbMATH: 07524956
Digital Object Identifier: 10.1214/21-EJS1975

Keywords: almost parametric rate of convergence , Assouad’s lemma , curse of dimensionality , minimax lower bounds , multivariate normal location mixtures

Vol.16 • No. 1 • 2022
Back to Top