Approximate Bayesian inference in semiparametric copula models

TL;DR

This paper introduces a straightforward Bayesian inference method for functionals of multivariate distributions using copula representations and an Approximate Bayesian Monte Carlo approach, especially useful when likelihood evaluation is costly or the model is partially specified.

Contribution

It presents a novel, computationally efficient Bayesian inference technique leveraging copulas and empirical likelihood, applicable to complex or partially specified models.

Findings

Effective in scenarios with costly likelihood evaluations

Applicable to partially specified models

Provides a practical alternative to traditional methods

Abstract

We describe a simple method for making inference on a functional of a multivariate distribution. The method is based on a copula representation of the multivariate distribution and it is based on the properties of an Approximate Bayesian Monte Carlo algorithm, where the proposed values of the functional of interest are weighed in terms of their empirical likelihood. This method is particularly useful when the "true" likelihood function associated with the working model is too costly to evaluate or when the working model is only partially specified.