Abstract

This paper discusses the use of Bayesian approaches when the models are misspecified (model risk). In particular we explore the limits and future of Bayesian approaches in order to provide answers to the following questions recently asked by the prudential supervision for Finance/Insurance: How to measure model risk? How to use in a coherent way the different misspecified models (as rating models) usually employed within and between financial institutions.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call