Some cautionary tales about Bayesian predictive inference

2026/07/21 by Emanuela Dreassi, Fabrizio Leisen, Luca Pratelli +1
#math.ST #stat.ME #stat.ML #stat.TH

paper · pdf

Abstract

Two misunderstandings, frequently arising in Bayesian predictive inference, are discussed. The first deals with the data generating mechanism, while the second consists in overestimating the role played by asymptotic exchangeability. Some consequences of such misunderstandings are highlighted through examples.

Related

• Exchangeability, prediction and predictive modeling in Bayesian statistics
• Bayesian inference: More than Bayes's theorem
• Bayesian Inference
• On the Stability of General Bayesian Inference
• Bayesian Predictive Inference Beyond Martingales