Boosting method models insurance claims frequency and severity jointly
Joint Distributional Copula Regression for Insurance Claim Frequency and Severity via Boosting
Machine Learning
Summary
Insurance companies want to understand how often claims happen and how much those claims cost. The authors developed a new statistical method that looks at both the number of claims and their costs together, using patterns in data. Their method is easy to interpret and helps find important factors influencing claims. It also performs well compared to other common techniques.
What this means in practice
- •For insurance analysts: Model the relationship between how often claims happen and their size to improve risk assessment.
- •For financial risk modelers: Use the joint modeling approach to better understand dependencies in loss data and improve forecasting accuracy.
Authors
Banghee So, Na Zhang
Abstract
In this paper, we propose a model-based boosting approach for fitting structured additive distributional copula regression models with bivariate responses, focusing particularly on the joint modeling of claim frequency and claim severity. The proposed approach is based on generalized additive models for location, scale, and shape (GAMLSS), which extend the general idea of GAMs by allowing multiple parameters of a response distribution, rather than only its conditional mean, to depend on covariates. The model is therefore intrinsically interpretable at the level of its additive predictors, providing greater transparency and auditability than black-box alternatives. We evaluate the proposed framework using synthetic and real-world data. We demonstrate its ability to recover the underlying covariate effects and dependence structure, identify informative variables, and achieve competitive predictive performance relative to benchmark models. We also illustrate how the fitted model can be interpreted.