|
Activity Number:
|
569
- Recent Advances in Predictive Inference
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 2, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #322725
|
View Presentation
|
|
Title:
|
On Sparse Bayes Predictive Density Estimates
|
|
Author(s):
|
Gourab Mukherjee*
|
|
Companies:
|
University of Southern California
|
|
Keywords:
|
Predictive density ;
Spike and Slab ;
Proper Bayes ;
Minimaxity ;
Sparsity
|
|
Abstract:
|
We study the problem of predictive density estimation under Kullback-Leibler loss in sparse Gaussian sequence models. Based on a bi-grid prior we propose a proper Bayes predictive density estimate which attains minimax optimality. The minimax risks of predictive density estimates based on popular spike and slab approaches are also studied. Comparing our proposed Bayes predictive density estimate with thresholding based minimax optimal rules, we explain new similarities and contrasts with the parallel theory of point estimation of a multivariate normal mean under quadratic loss.
|
Authors who are presenting talks have a * after their name.
