The paper proposes a ratemaking framework for claim frequency that uses informative telematics data and complies with a “discount-only” regulatory requirement of the sort proposed in the 2023–2024 session of the New York State Assembly. The proposed framework uses a feedforward neural network to extract a one-dimensional safety score from multidimensional telematics features and integrates that score with traditional features in generalized linear models (GLMs). To meet the discount-only requirement, we impose constraints on the safety score and its regression parameter. The results show that the proposed models, with a suitable safety score function, can outperform a standard GLM in both in-sample goodness of fit and out-of-sample prediction performance. Furthermore, the analysis reveals that while the discount-only constraint may drive insurers to raise base premiums to offset revenue losses from the relativity cap, the regulation could achieve its intended goal in scenarios with strong favorable selection. This work was supported by a 2024 Individual Research Grant from the Casualty Actuarial Society.
Peiris et al. (Tue,) studied this question.