Tweedie Gradient Boosting for Extremely Unbalanced Zero-inflated Data

TL;DR

This paper introduces EMTboost, a novel gradient boosting method for zero-inflated Tweedie models, effectively handling extremely unbalanced insurance claim data with complex predictor interactions.

Contribution

It develops a zero-inflated Tweedie boosting model with a nonparametric component and a specialized EM algorithm, improving modeling of zero-inflated, skewed data.

Findings

Enhanced prediction accuracy on synthetic auto-insurance data

Effectively captures nonlinearities and complex interactions

Outperforms traditional Tweedie models in zero-inflated scenarios

Abstract

Tweedie's compound Poisson model is a popular method to model insurance claims with probability mass at zero and nonnegative, highly right-skewed distribution. In particular, it is not uncommon to have extremely unbalanced data with excessively large proportion of zero claims, and even traditional Tweedie model may not be satisfactory for fitting the data. In this paper, we propose a boosting-assisted zero-inflated Tweedie model, called EMTboost, that allows zero probability mass to exceed a traditional model. We makes a nonparametric assumption on its Tweedie model component, that unlike a linear model, is able to capture nonlinearities, discontinuities, and complex higher order interactions among predictors. A specialized Expectation-Maximization algorithm is developed that integrates a blockwise coordinate descent strategy and a gradient tree-boosting algorithm to estimate key model parameters. We use extensive simulation and data analysis on synthetic zero-inflated auto-insurance claim data to illustrate our method's prediction performance.