CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Robustness in Bayesian inference Authors:  Laura Ventura - University of Padova (Italy) [presenting]
Erlis Ruli - University of Padova (Italy)
Nicola Sartori - University of Padova (Italy)
Abstract: The aim is to review the properties and applications of the so-called robust posterior distributions, i.e. posterior distributions derived from the combination of a robust pseudo-likelihood function or an unbiased estimating function with suitable prior information. Examples of pseudo-likelihoods are the composite, the empirical and the quasi-likelihoods, while unbiased estimating functions include as special instances M-estimating functions and proper scoring rules. From a theoretical point of view we illustrate how to perform robust Bayesian inference from pseudo-likelihoods and estimating equations. From a practical point of view, we show the simple but effective application of robust posterior distributions in challenging examples.