English

Non-Bayesian Learning in Misspecified Models

Theoretical Economics 2025-10-06 v3 Statistics Theory Statistics Theory

Abstract

Deviations from Bayesian updating are traditionally categorized as biases, errors, or fallacies, thus implying their inherent ``sub-optimality.'' We offer a more nuanced view. We demonstrate that, in learning problems with misspecified models, non-Bayesian updating can outperform Bayesian updating.

Cite

@article{arxiv.2503.18024,
  title  = {Non-Bayesian Learning in Misspecified Models},
  author = {Sebastian Bervoets and Mathieu Faure and Ludovic Renou},
  journal= {arXiv preprint arXiv:2503.18024},
  year   = {2025}
}
R2 v1 2026-06-28T22:31:17.382Z