Date: 2023-02-16

Time: 14:00-15:00 (UK time)

Strand S0.13

Abstract

De Finetti promoted the importance of predictive models for observables as the basis for Bayesian inference. The assumption of exchangeability, implying aspects of symmetry in the predictive model, motivates the usual likelihood-prior construction and with it the traditional learning approach involving a prior to posterior update using Bayes’ rule. We discuss an alternative approach, treating Bayesian inference as a missing data problem for observables not yet obtained from the population needed to estimate a parameter precisely or make a decision correctly. This motivates the direct use of predictive models for inference, relaxing exchangeability to start modelling from the data in hand (with or without a prior). Martingales play a key role in the construction. This is joint work with Stephen Walker and Edwin Fong, based on the paper “Martingale Posteriors” to appear with discussion JRSS Series B.

Speaker

Chris Holmes hold a joint professorship in Biostatistics at the University of Oxford with a joint appointment between Departments of Statistics and the Nuffield Department of Medicine at the University of Oxford. He is also is the Programme Director for Health and Medical Sciences at the Alan Turing Institute. His research explores the potential of computational statistics and statistical machine learning to assist in the medical and health sciences. He is particularly interested in pattern recognition and nonlinear, nonparametric statistical machine learning methods applied to the genomic sciences and genetic epidemiology.