Recalibration: A post-processing method for approximate Bayesian computation

TL;DR

This paper introduces a recalibration post-processing method to enhance the accuracy of posterior approximations in Approximate Bayesian Computation, compatible with existing methods and leveraging auxiliary models for improved inference.

Contribution

It proposes a novel recalibration technique for ABC that improves posterior quality and extends the use of misspecified auxiliary models, with implementation in an R package.

Findings

Recalibration improves posterior approximation accuracy in simulated examples.

The method enhances ABC analysis of stereological extremes.

Recalibration is compatible with existing post-processing techniques.

Abstract

A new recalibration post-processing method is presented to improve the quality of the posterior approximation when using Approximate Bayesian Computation (ABC) algorithms. Recalibration may be used in conjunction with existing post-processing methods, such as regression-adjustments. In addition, this work extends and strengthens the links between ABC and indirect inference algorithms, allowing more extensive use of misspecified auxiliary models in the ABC context. The method is illustrated using simulated examples to demonstrate the effects of recalibration under various conditions, and through an application to an analysis of stereological extremes both with and without the use of auxiliary models. Code to implement recalibration post-processing is available in the R package, abctools.