Open Access
Translator Disclaimer
May 2021 Minimax predictive density for sparse count data
Keisuke Yano, Ryoya Kaneko, Fumiyasu Komaki
Author Affiliations +
Bernoulli 27(2): 1212-1238 (May 2021). DOI: 10.3150/20-BEJ1271

Abstract

This paper discusses predictive densities under the Kullback–Leibler loss for high-dimensional Poisson sequence models under sparsity constraints. Sparsity in count data implies zero-inflation. We present a class of Bayes predictive densities that attain asymptotic minimaxity in sparse Poisson sequence models. We also show that our class with an estimator of unknown sparsity level plugged-in is adaptive in the asymptotically minimax sense. For application, we extend our results to settings with quasi-sparsity and with missing-completely-at-random observations. The simulation studies as well as application to real data illustrate the efficiency of the proposed Bayes predictive densities.

Citation

Download Citation

Keisuke Yano. Ryoya Kaneko. Fumiyasu Komaki. "Minimax predictive density for sparse count data." Bernoulli 27 (2) 1212 - 1238, May 2021. https://doi.org/10.3150/20-BEJ1271

Information

Received: 1 November 2019; Revised: 1 June 2020; Published: May 2021
First available in Project Euclid: 24 March 2021

Digital Object Identifier: 10.3150/20-BEJ1271

Keywords: Adaptation , high dimension , Kullback–Leibler divergence , missing at random , Poisson model , Zero inflation

Rights: Copyright © 2021 ISI/BS

JOURNAL ARTICLE
27 PAGES


SHARE
Vol.27 • No. 2 • May 2021
Back to Top