Webinars

Individual loss reserving methods have undergone substantial development in the past decade, driven by increased access to granular-level insurance claims data. This research aims to propose a micro loss reserving model tailored for multi-coverage insurance policies, where a single insurance claim might trigger payments from multiple coverage types. A copula-based multivariate regression approach is employed to jointly model the settlement time and loss amount, effectively capturing the dependence among various types of loss amounts and their correlation with the settlement time. We stress the importance of considering both types of dependence for accurate reserving prediction and uncertainty quantification. Furthermore, we propose computationally efficient algorithms for parameter estimation and dynamic prediction. Through numerical experiments and real data analysis, we demonstrate the effectiveness of the proposed multivariate predictive model in loss reserving applications.

Predicting and understanding solvency figures poses major challenges for life insurers. Internal risk models provide realistic representations, but they remain company-internal and make model understanding difficult. This makes it hard to compare new machine-learning methods for prediction and explainability objectively and under realistic scenarios.

We developed openIRM, an open-source internal risk model for an artificial life insurer. It combines an economic scenario generator based on the G2++ interest-rate model with a cash-flow projection model. The model supports both outer real-world simulations and inner simulations for determining basic own funds. It has also been calibrated for all trading days from September 2016 to December 2023 and allows the use of direct and indirect valuation methods, which have been shown to produce consistent results.

This talk gives a short introduction to risk modelling for a life insurer under Solvency II, outlines how the openIRM model works, and demonstrates its practical use through a concrete example.