Explainable AI

This paper explores the tuning and results of two-part models on rich datasets provided through the Casualty Actuarial Society (CAS). …

mSHAP values can be used to gain insights on variable importance when using two-part models that are commonly used in insurance modeling.

Summary Presented my work on MSHAP values at the 24th International Insurance: Mathematics and Economics Conference in July of 2021. This conference was held in a virtual format.