City Research Online

Abstract

The generalised linear model is a flexible predictive model for observational data that is widely used in practice as it extends linear regression models to non-Gaussian data. In this paper we introduce the concept of a properly defined generalised linear model by requiring the conditional mean of the response variable to be properly mapped through the chosen link function and the log-likelihood function to be concave. We provide a comprehensive classification of proper generalised linear models for the Tweedie family and its popular subclasses under different link function specifications. Our main theoretical findings show that most Tweedie generalised linear models are not proper for canonical and log link functions, and identify a rich class of proper Tweedie generalised linear models with power link functions. Using self-concordant log-likelihoods and linearisation techniques, we provide novel algorithms for estimating several special cases of proper and not proper Tweedie generalised linear models with power link functions. The effectivenes of our methods is determined through an extensive numerical comparison of our estimates and those obtained using three built-in packages, MATLAB f itglm, R glm2 and Python sm.GLM libraries, which are all implemented based on the standard Iteratively Reweighted Least Squares method. Overall, we find that our algorithms consistently outperform these benchmarks in terms of both accuracy and efficiency, the largest improvements being documented for high-dimensional settings.

Publication Type:	Other (Preprint)
Additional Information:	Copyright the authors, 2024.
Publisher Keywords:	Exponential dispersion family, proper generalised linear model, Tweedie regression, power link function, self-concordance
Subjects:	H Social Sciences > HF Commerce Q Science > QA Mathematics
Departments:	Bayes Business School > Actuarial Science & Insurance

Export

Downloads