The expanding application of advanced analytics in insurance has generated numerous opportunities, such as more accurate predictive modelling powered by Machine Learning and Artificial Intelligence methods, the utilization of novel and unstructured datasets, and the automation of key operations. Significant advances in these areas are being made through novel applications and adaptations of predictive modelling techniques for insurance purposes, while, concurrently, rapid advances in machine learning methods are being made outside of the insurance sector. At the same time, substantial challenges arise revolving around the transparency of complex algorithmic models and the economic and societal impacts of their adoption in decision making. As insurance is a highly regulated industry, models may be required by regulators to be explainable, in order to enable analysis of the basis for decision making. Due to the societal importance of insurance, significant attention is being paid to ensuring that insurance models do not discriminate unfairly.

Contributions are invited for a special issue of the Annals of Actuarial Science, to feature research on ‘Insurance analytics: from prediction to explainability’. Areas of interest include, but are not limited to:

• Methodological innovations in predictive analytics applied in insurance
• Case studies of applications of machine learning and artificial intelligence within insurance, where techniques have been adapted significantly to the insurance context or illustrating use of novel datasets
• Methods for interpreting predictive models and designing inherently explainable models as applied in an insurance context
• Addressing discrimination and fairness considerations in insurance pricing and related applications
• Algorithmic auditing and validation of predictive models in insurance
• Software that enables the operationalisation of methods relating to the scope of the special issue, e.g. R/Python packages, dashboards, novel open source datasets etc.
• New developments with insurance applications in areas closely related to actuarial science, such as finance, bio-statistics and demography

This special issue aims to capture leading academic thinking and industry ap- plications in advanced insurance analytics, encompassing long-established (e.g. pricing and reserving) to novel (e.g. claims fraud) areas of application. Submissions should illustrate advances in at least one (and potentially all) of the domains of predictive accuracy, explainable modelling and fairness, while reflecting on recent progress already made in these areas. Since applying machine learning methods within insurance often requires adaptation of techniques used more generally, submissions should maintain focus on how this has been done. 

The Guest Editors of this special issue are:

• Kjersti Aas, Norwegian Computing Center & Norwegian University of Science and Technology
• Arthur Charpentier, Université du Québec à Montréal Canada
• Fei Huang, University of New South Wales
• Ronald Richman, Old Mutual Insure & University of the Witwatersrand

Papers intended for the special issue may be submitted from 1 December 2022 to 30 September 2023. The special issue will appear in 2024, but all papers accepted for publication will be published as FirstView articles when produced. 

Upon submission via ScholarOne, please indicate that your manuscript is intended for this special issue by selecting the appropriate option from the dropdown menu in the Special Issue section. The journal’s usual submission instructions for original research and review articles and for software contributions apply. For further detail please see the journal’s information pages, or contact [email protected]. Questions regarding the scope of the special issue can be addressed to the Editor-in-Chief, Andreas Tsanakas.