Asymptotics of cut distributions and robust modular inference using Posterior Bootstrap

10/21/2021
by   Emilia Pompe, et al.
0

Bayesian inference provides a framework to combine an arbitrary number of model components with shared parameters, allowing joint uncertainty estimation and the use of all available data sources. However, misspecification of any part of the model might propagate to all other parts and lead to unsatisfactory results. Cut distributions have been proposed as a remedy, where the information is prevented from flowing along certain directions. We consider cut distributions from an asymptotic perspective, find the equivalent of the Laplace approximation, and notice a lack of frequentist coverage for the associate credible regions. We propose algorithms based on the Posterior Bootstrap that deliver credible regions with the nominal frequentist asymptotic coverage. The algorithms involve numerical optimization programs that can be performed fully in parallel. The results and methods are illustrated in various settings, such as causal inference with propensity scores and epidemiological studies.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/21/2022

Cutting feedback and modularized analyses in generalized Bayesian inference

Even in relatively simple settings, model misspecification can make the ...
research
08/25/2021

Variational inference for cutting feedback in misspecified models

Bayesian analyses combine information represented by different terms in ...
research
01/30/2022

Causal inference under mis-specification: adjustment based on the propensity score

We study Bayesian approaches to causal inference via propensity score re...
research
03/18/2022

Modularized Bayesian analyses and cutting feedback in likelihood-free inference

There has been much recent interest in modifying Bayesian inference for ...
research
01/26/2023

Guaranteed Accuracy of Semi-Modular Posteriors

Bayesian inference has widely acknowledged advantages in many problems, ...
research
04/12/2022

Specifying Prior Distributions in Reliability Applications

Especially when facing reliability data with limited information (e.g., ...
research
02/07/2020

Random weighting to approximate posterior inference in LASSO regression

We consider a general-purpose approximation approach to Bayesian inferen...

Please sign up or login with your details

Forgot password? Click here to reset