Martingale Posterior Distributions

by   Edwin Fong, et al.

The prior distribution on parameters of a likelihood is the usual starting point for Bayesian uncertainty quantification. In this paper, we present a different perspective. Given a finite data sample Y_1:n of size n from an infinite population, we focus on the missing Y_n+1:∞ as the source of statistical uncertainty, with the parameter of interest being known precisely given Y_1:∞. We argue that the foundation of Bayesian inference is to assign a predictive distribution on Y_n+1:∞ conditional on Y_1:n, which then induces a distribution on the parameter of interest. Demonstrating an application of martingales, Doob shows that choosing the Bayesian predictive distribution returns the conventional posterior as the distribution of the parameter. Taking this as our cue, we relax the predictive machine, avoiding the need for the predictive to be derived solely from the usual prior to posterior to predictive density formula. We introduce the martingale posterior distribution, which returns Bayesian uncertainty directly on any statistic of interest without the need for the likelihood and prior, and this distribution can be sampled through a computational scheme we name predictive resampling. To that end, we introduce new predictive methodologies for multivariate density estimation, regression and classification that build upon recent work on bivariate copulas.



There are no comments yet.


page 23

page 27


Bayesian inference and uncertainty quantification for image reconstruction with Poisson data

We provide a complete framework for performing infinite-dimensional Baye...

The Bayesian Method of Tensor Networks

Bayesian learning is a powerful learning framework which combines the ex...

Bayesian predictive inference without a prior

Let (X_n:n≥ 1) be a sequence of random observations. Let σ_n(·)=P(X_n+1∈...

Uncertainty Quantification of the 4th kind; optimal posterior accuracy-uncertainty tradeoff with the minimum enclosing ball

There are essentially three kinds of approaches to Uncertainty Quantific...

Bayesian Conditional Density Filtering

We propose a Conditional Density Filtering (C-DF) algorithm for efficien...

Bayesian estimation of probabilistic sensitivity measures

Computer experiments are becoming increasingly important in scientific i...

A numerically stable online implementation and exploration of WAIC through variations of the predictive density, using NIMBLE

We go through the process of crafting a robust and numerically stable on...

Code Repositories


Martingale Posteriors with Copulas

view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.