Local posterior concentration rate for multilevel sparse sequences

E.N. Belitser, N. Nurushev

Research output: Chapter in Book / Report / Conference proceedingChapterAcademicpeer-review

Abstract

We consider empirical Bayesian inference in the many normal means
model in the situation when the high-dimensional mean vector is multilevel sparse, that is,most of the entries of the parameter vector are some fixed values. For instance, the traditional sparse signal is a particular case (with one level) of multilevel sparse sequences.We apply an empirical Bayesian approach, namely we put an appropriate prior modeling the multilevel sparsity and make data-dependent choices of certain parameters of the prior. We establish local (i.e., with rate depending on the “true” parameter) posterior contraction and estimation results. Global adaptive minimax results (for the estimation and posterior contraction problems) over sparsity classesfollow from our local results if the sparsity level is of polynomial order. The results are illustrated by simulations.
Original languageEnglish
Title of host publicationBayesian Statistics in Action
PublisherSpringer
Pages51-66
Number of pages16
Volume194
ISBN (Electronic) 978-3-319-54084-9
ISBN (Print)978-3-319-54083-2
DOIs
Publication statusPublished - 2017

Publication series

NameSpringer Proceedings in Mathematics & Statistics
PublisherSpringer

Keywords

  • Empirical Bayesian approach
  • Local posterior concentration rate
  • Multilevel sparse sequences

Fingerprint Dive into the research topics of 'Local posterior concentration rate for multilevel sparse sequences'. Together they form a unique fingerprint.

Cite this