Ambiguity Types, Robust Learning and Natural Catastrophe Insurance: How Long-Term Contracts May Help

Abstract

Motivated by the results of the field experiment in the United Sates to distinguish two sources of ambiguity and its relation with the robust learning theory, we propose an insurance pricing formula to accommodate the ambiguity types in the robust learning framework. Based on the field experiment results and the data of the yield spread of catastrophe linked securities as well as their expected loss, our empirical test separates the magnitudes of different types of ambiguity aversion over different times for different periods. A related four-period model is then established to discuss long-term insurance (LTI) as an alternative to the standard annual insurance policy.

Keywords: Ambiguity Types, Robust Learning, Insurance Pricing, Esscher Transform, LTI

JEL Classification: C93, D81, D83