Media mSHAP: Explainability of Two-Part Models

mSHAP: Explainability of Two-Part Models

uploaded August 3, 2021 Views: 5 Comments: 0 Marked favorite: 0

Covariate inference in two-part models composed of black box algorithms is a difficult task. Recent breakthroughs in explainable machine learning as a whole have paved the way to improve this explainability, but little progress has been made in two-part models. In this paper, we propose mSHAP (or multiplicative SHAP), a method for computing SHAP values of two-part models using the SHAP values of the individual models. This method will allow for the predictions of two-part models to be explained at an individual observation level. After developing mSHAP, we perform an in-depth simulation study.  Although the kernelSHAP algorithm is also capable of computing approximate SHAP values for a two-part model, a comparison with our method demonstrates that mSHAP is exponentially faster. Ultimately, we apply mSHAP to a two-part ratemaking model for personal auto property damage.

Content groups:  content2021


There are no comments yet. Add a comment.