Sample Size Determination for Credibility Estimation

Abstract

The credibility estimator has been successfully and widely applied by practitioners in the insurance industry for decades. However, an important issue remains unresolved: what sample size should the actuary take when he/she applies the credibility estimator? In particular, is the size of a given claims data set large enough for the credibility estimator to be accurate? This paper aims to address this issue by first suggesting a sample size criterion for the credibility estimator and then proposing a sample size based upon that. The proposed sample size is easy to apply, and it guarantees provably accuracy for credibility estimation.

Keywords: Bayesian analysis; credibility theory; experience rating; predictive modeling; sample size criterion

JEL Classification: G22