Explainable Business Analytics for Usage-Based Auto Insurance: GAM, SHAP, and Territory Risk Segmentation in Telematics Pricing
DOI:
https://doi.org/10.63646/AGLR1659Keywords:
Usage-based insurance; telematics analytics; generalized additive models; explainable ai; shap; interpretable clustering; territory risk; rate regulationAbstract
The shift from group-based to individual-based pricing in personal auto insurance has been accelerated by the diffusion of telematics and usage-based insurance (UBI) programmes. Although UBI portfolios are rich in behavioural signals, regulators still demand pricing models that are statistically transparent and that allow each rating relativity to be defended. This study develops a unified explainable business analytics framework that integrates Generalized Linear Models (GLM), Generalized Additive Models (GAM), gradient-boosted trees, Shapley Additive exPlanations (SHAP) and an interpretable two-dimensional K-means territory clustering procedure. The framework is applied to a synthetic UBI portfolio that replicates the distributional structure of a Canadian transactional dataset, with claim frequency and claim severity modelled separately under appropriate exponential-family distributions. Group-level RMSE shows that GAMs reduce error by roughly 33 percent relative to GLMs across the five core rating dimensions, and they better recover the curvature of risk as a function of annual mileage, credit score and years without claims. Variable-importance evidence from SHAP applied to an XGBoost model confirms credit score, years without claims and car age as the dominant frequency drivers, while annual miles driven becomes most informative once interactions with car use and region are admitted. A regularized cluster-selection criterion that augments mean absolute deviation with a complexity penalty selects between 11 and 14 territory groups, generating compact, monotonic and policy-defensible territory bands. Sensitivity analysis over the regularization parameter alpha confirms the stability of the selected K. The results provide a reproducible blueprint for regulators and insurers seeking to combine predictive performance with the interpretability required by rate-filing reviews.
How to Cite
References
Ayuso, M., Guillen, M., & Pérez-Marín, A. M. (2014). Time and distance to first accident and driving patterns of young drivers with pay-as-you-drive insurance. Accident Analysis & Prevention, 73, 125-131. https://doi.org/10.1016/j.aap.2014.08.017
Ayuso, M., Guillen, M., & Pérez-Marín, A. M. (2016). Telematics and gender discrimination: Some usage-based evidence on whether men's risk of accidents differs from women's. Risks, 4(2), 10. https://doi.org/10.3390/risks4020010
Boucher, J.-P., Côté, S., & Guillen, M. (2017). Exposure as duration and distance in telematics motor insurance using generalized additive models. Risks, 5(4), 54. https://doi.org/10.3390/risks5040054
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324
Cevolini, A., & Esposito, E. (2020). From pool to profile: Social consequences of algorithmic prediction in insurance. Big Data & Society, 7(2), 1-11. https://doi.org/10.1177/2053951720939228
Charpentier, A. (2014). Computational Actuarial Science with R. CRC Press. https://doi.org/10.1201/b17230
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794. https://doi.org/10.1145/2939672.2939785
Eling, M., & Kraft, M. (2020). The impact of telematics on the insurability of risks. The Journal of Risk Finance, 21(2), 77-109. https://doi.org/10.1108/JRF-07-2019-0129
Eling, M., & Lehmann, M. (2018). The impact of digitalization on the insurance value chain and the insurability of risks. The Geneva Papers on Risk and Insurance - Issues and Practice, 43(3), 359-396. https://doi.org/10.1057/s41288-017-0073-0
Frees, E. W., & Huang, F. (2023). The discriminating (pricing) actuary. North American Actuarial Journal, 27(1), 2-24. https://doi.org/10.1080/10920277.2021.1951296
Frees, E. W., Derrig, R. A., & Meyers, G. (Eds.). (2014). Predictive Modeling Applications in Actuarial Science. Cambridge University Press. https://doi.org/10.1017/CBO9781139342674
Frees, E. W., Meyers, G., & Cummings, A. D. (2011). Summarizing insurance scores using a Gini index. Journal of the American Statistical Association, 106(495), 1085-1098. https://doi.org/10.1198/jasa.2011.tm10506
Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29(5), 1189-1232. https://doi.org/10.1214/aos/1013203451
Friedman, J. H. (2002). Stochastic gradient boosting. Computational Statistics & Data Analysis, 38(4), 367-378. https://doi.org/10.1016/S0167-9473(01)00065-2
Gao, G., & Wuthrich, M. V. (2018). Feature extraction from telematics car driving heatmaps. European Actuarial Journal, 8(2), 383-406. https://doi.org/10.1007/s13385-018-0181-7
Goldburd, M., Khare, A., Tevet, D., & Guller, D. (2016). Generalized linear models for insurance rating. CAS Monograph Series, 5. https://www.casact.org/sites/default/files/2021-01/05-Goldburd-Khare-Tevet.pdf
Guillen, M., Nielsen, J. P., Ayuso, M., & Pérez-Marín, A. M. (2019). The use of telematics devices to improve automobile insurance rates. Risk Analysis, 39(3), 662-672. https://doi.org/10.1111/risa.13172
Hartigan, J. A., & Wong, M. A. (1979). Algorithm AS 136: A K-means clustering algorithm. Journal of the Royal Statistical Society. Series C (Applied Statistics), 28(1), 100-108. https://doi.org/10.2307/2346830
Hastie, T., & Tibshirani, R. (1986). Generalized additive models. Statistical Science, 1(3), 297-310. https://doi.org/10.1214/ss/1177013604
Henckaerts, R., Antonio, K., Clijsters, M., & Verbelen, R. (2018). A data driven binning strategy for the construction of insurance tariff classes. Scandinavian Actuarial Journal, 2018(8), 681-705. https://doi.org/10.1080/03461238.2018.1429300
Henckaerts, R., Côté, M.-P., Antonio, K., & Verbelen, R. (2021). Boosting insights in insurance tariff plans with tree- based machine learning methods. North American Actuarial Journal, 25(2), 255-285. https://doi.org/10.1080/10920277.2020.1745656
Husnjak, S., Peraković, D., Forenbacher, I., & Mumdziev, M. (2015). Telematics system in usage based motor insurance. Procedia Engineering, 100, 816-825. https://doi.org/10.1016/j.proeng.2015.01.436
Jolliffe, I. T. (2002). Principal Component Analysis (2nd ed.). Springer. https://doi.org/10.1007/b98835
Klein, N., Denuit, M., Lang, S., & Kneib, T. (2014). Nonlife ratemaking and risk management with Bayesian generalized additive models for location, scale, and shape. Insurance: Mathematics and Economics, 55, 225-249. https://doi.org/10.1016/j.insmatheco.2014.02.001
Lemaire, J., Park, S. C., & Wang, K. C. (2016). The use of annual mileage as a rating variable. ASTIN Bulletin: The Journal of the IAA, 46(1), 39-69. https://doi.org/10.1017/asb.2015.25
Lindholm, M., Richman, R., Tsanakas, A., & Wuthrich, M. V. (2022). Discrimination-free insurance pricing. ASTIN Bulletin, 52(1), 55-89. https://doi.org/10.1017/asb.2021.23
Lloyd, S. (1982). Least squares quantization in PCM. IEEE Transactions on Information Theory, 28(2), 129-137. https://doi.org/10.1109/TIT.1982.1056489
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. NeurIPS, 30. https://papers.nips.cc/paper/7062-a-unified-approach-to-interpreting-model-predictions
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S.-I. (2020). From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence, 2(1), 56-67. https://doi.org/10.1038/s42256-019-0138-9
McCullagh, P., & Nelder, J. A. (2019). Generalized Linear Models (2nd ed., reprint). Chapman and Hall/CRC. https://doi.org/10.1201/9780203753736
Molnar, C. (2020). Interpretable machine learning: A guide for making black box models explainable. Lulu.com. ISBN: 9780244768522.
Nelder, J. A., & Wedderburn, R. W. M. (1972). Generalized linear models. Journal of the Royal Statistical Society. Series A, 135(3), 370-384. https://doi.org/10.2307/2344614
Pesantez-Narvaez, J., Guillen, M., & Alcañiz, M. (2019). Predicting motor insurance claims using telematics data: XGBoost versus logistic regression. Risks, 7(2), 70. https://doi.org/10.3390/risks7020070
Pérez-Marín, A. M., Guillen, M., Alcañiz, M., & Bermúdez, L. (2019). Quantile regression with telematics information to assess the risk of driving above the posted speed limit. Risks, 7(3), 80. https://doi.org/10.3390/risks7030080
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). "Why should I trust you?": Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135- 1144. https://doi.org/10.1145/2939672.2939778
Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53-65. https://doi.org/10.1016/0377-0427(87)90125-7
Siami, M., Naderpour, M., & Lu, J. (2020). A mobile telematics pattern recognition framework for driving behavior extraction. IEEE Transactions on Intelligent Transportation Systems, 22(3), 1459-1472. https://doi.org/10.1109/TITS.2020.2971214
So, B., Boucher, J.-P., & Valdez, E. A. (2021). Synthetic dataset generation of driver telematics. Risks, 9(4), 58. https://doi.org/10.3390/risks9040058
Stipancic, J., Miranda-Moreno, L., & Saunier, N. (2018). Vehicle manoeuvers as surrogate safety measures: Extracting data from the GPS-enabled smartphones of regular drivers. Accident Analysis & Prevention, 115, 160-169. https://doi.org/10.1016/j.aap.2018.03.005
Strumbelj, E., & Kononenko, I. (2014). Explaining prediction models and individual predictions with feature contributions. Knowledge and Information Systems, 41(3), 647-665. https://doi.org/10.1007/s10115-013-0679-x
Sugar, C. A., & James, G. M. (2003). Finding the number of clusters in a dataset: An information-theoretic approach. Journal of the American Statistical Association, 98(463), 750-763. https://doi.org/10.1198/016214503000000666
Sun, S., Bi, J., Guillen, M., & Pérez-Marín, A. M. (2020). Assessing driving risk using Internet of Vehicles data: An analysis based on generalized linear models. Sensors, 20(9), 2712. https://doi.org/10.3390/s20092712
Sundararajan, M., Taly, A., & Yan, Q. (2017). Axiomatic attribution for deep networks. Proceedings of the 34th International Conference on Machine Learning, 70, 3319–3328. https://proceedings.mlr.press/v70/sundararajan17a.html
Tibshirani, R., Walther, G., & Hastie, T. (2001). Estimating the number of clusters in a data set via the gap statistic. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 63(2), 411-423. https://doi.org/10.1111/1467-9868.00293
Verbelen, R., Antonio, K., & Claeskens, G. (2018). Unravelling the predictive power of telematics data in car insurance pricing. Journal of the Royal Statistical Society. Series C (Applied Statistics), 67(5), 1275-1304. https://doi.org/10.1111/rssc.12283
Weidner, W., Transchel, F. W. G., & Weidner, R. (2017). Telematic driving profile classification in car insurance pricing. Annals of Actuarial Science, 11(2), 213-236. https://doi.org/10.1017/S1748499516000130
Wood, S. N. (2003). Thin plate regression splines. Journal of the Royal Statistical Society: Series B, 65(1), 95-114. https://doi.org/10.1111/1467-9868.00374
Wood, S. N. (2004). Stable and efficient multiple smoothing parameter estimation for generalized additive models. Journal of the American Statistical Association, 99(467), 673-686. https://doi.org/10.1198/016214504000000980
Wood, S. N. (2017). Generalized Additive Models: An Introduction with R (2nd ed.). Chapman and Hall/CRC. https://doi.org/10.1201/9781315370279
Wuthrich, M. V. (2017). Covariate selection from telematics car driving data. European Actuarial Journal, 7(1), 89-108. https://doi.org/10.1007/s13385-017-0149-z
Xie, S., & Gan, C. (2022). Fuzzy clustering and non-negative sparse matrix approximation on estimation of territorial auto insurance risk relativities. Soft Computing, 26(20), 11187-11200. https://doi.org/10.1007/s00500-022-07378-0
Xie, S., & Lawniczak, A. T. (2018). Estimating major risk factor relativities in rate filings using generalized linear models. International Journal of Financial Studies, 6(4), 84. https://doi.org/10.3390/ijfs6040084
Yip, K. C. H., & Yau, K. K. W. (2005). On modeling claim frequency data in general insurance with extra zeros. Insurance: Mathematics and Economics, 36(2), 153-163. https://doi.org/10.1016/j.insmatheco.2004.11.002