From Point to probabilistic gradient boosting for claim frequency and severity prediction

Abstract

Gradient boosting for decision tree algorithms are increasingly used in actuarial applications as they show superior predictive performance over traditional generalised linear models. Many enhancements to the first gradient boosting machine algorithm exist. We present in a unified notation, and contrast, all the existing point and probabilistic gradient boosting for decision tree algorithms: GBM, XGBoost, DART, LightGBM, CatBoost, EGBM, PGBM, XGBoostLSS, cyclic GBM, and NGBoost. In this comprehensive numerical study, we compare their performance on five publicly available datasets for claim frequency and severity, of various sizes and comprising different numbers of (high cardinality) categorical variables. We explain how varying exposure-to-risk can be handled with boosting in frequency models. We compare the algorithms on the basis of computational efficiency, predictive performance, and model adequacy. LightGBM and XGBoostLSS win in terms of computational efficiency. CatBoost sometimes improves predictive performance, especially in the presence of high cardinality categorical variables, common in actuarial science. The fully interpretable EGBM achieves competitive predictive performance compared to the black box algorithms considered. We find that there is no trade-off between model adequacy and predictive accuracy: both are achievable simultaneously.

arXiv comment39 pages, 12 figures, 25 tables, 7 algorithms

References in corpus (10)

• XGBoost: A Scalable Tree Boosting System
• Predictive learning via rule ensembles
• InterpretML: A Unified Framework for Machine Learning Interpretability
• Probabilistic Gradient Boosting Machines for Large-Scale Probabilistic Regression
• Model Comparison and Calibration Assessment: User Guide for Consistent Scoring Functions in Machine Learning and Actuarial Practice
• CatBoostLSS -- An extension of CatBoost to probabilistic forecasting
• Neural networks for insurance pricing with frequency and severity data: a benchmark study from data preprocessing to technical tariff
• Double Probability Integral Transform Residuals for Regression Models with Discrete Outcomes
• Why You Should Not Trust Interpretations in Machine Learning: Adversarial Attacks on Partial Dependence Plots
• Zero-Inflated Tweedie Boosted Trees with CatBoost for Insurance Loss Analytics