Open Access
March 2012 Universality of Bayesian Predictions
Alessio Sancetta
Bayesian Anal. 7(1): 1-36 (March 2012). DOI: 10.1214/12-BA701

Abstract

This paper studies the theoretical properties of Bayesian predictions and shows that under minimal conditions we can derive finite sample bounds for the loss incurred using Bayesian predictions under the Kullback-Leibler divergence. In particular, the concept of universality of predictions is discussed and universality is established for Bayesian predictions in a variety of settings. These include predictions under almost arbitrary loss functions, model averaging, predictions in a non-stationary environment and under model misspecification.

Citation

Download Citation

Alessio Sancetta. "Universality of Bayesian Predictions." Bayesian Anal. 7 (1) 1 - 36, March 2012. https://doi.org/10.1214/12-BA701

Information

Published: March 2012
First available in Project Euclid: 13 June 2012

zbMATH: 1330.62151
MathSciNet: MR2896708
Digital Object Identifier: 10.1214/12-BA701

Keywords: Bayesian methods , loss function , model averaging , structural change , universal prediction

Rights: Copyright © 2012 International Society for Bayesian Analysis

Vol.7 • No. 1 • March 2012
Back to Top