Adapting the insurance pricing model for distribution channel expansion using the Bayesian generalized linear model

• Autorzy: Gunawan Carina , Faizal Muhammad Ivan , Susyanto Nanang

Identyfikatory

DOI

10.37190/ord240404

• Języki publikacji: EN

Abstrakty

EN

The insurance market is changing due to new distribution channels, requiring insurers to update their pricing models. We propose a mathematical approach using Bayesian generalized linear models (GLM) to adjust insurance pricing. Our strategy modifies the pricing model by incorporating distribution channels while utilizing the initial model as a baseline. Bayesian GLM enable effective model updates while incorporating existing knowledge. We validated our approach using data from the general insurance sector, comparing it with the traditional approach. Results show that Bayesian GLM outperforms the traditional method in accurately estimating pricing. This superiority highlights its potential as a powerful tool for insurers to remain competitive in a rapidly changing market. Our approach makes a significant mathematical contribution to insurance pricing, allowing insurers to adapt to market conditions and enhance their competitive edge.

Słowa kluczowe

EN

distribution channel; insurance pricing; competitiveness; predictive performance; Bayesian GLM

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Operations Research and Decisions
• Rocznik: 2024
• Tom: Vol. 34, No. 4
• Strony: 67--79
• Opis fizyczny: Bibliogr. 20 poz., rys.

Twórcy

autor

Gunawan Carina

• Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada, Yogyakarta, Indonesia

autor

Faizal Muhammad Ivan

• Asuransi Tugu Pratama Indonesia Tbk., Jakarta, Indonesia

autor

Susyanto Nanang

• Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada, Yogyakarta, Indonesia

• https://orcid.org/0000-0001-8332-3363

Bibliografia

• [1] Brooks, S., Gelman, A., Jones, G. L., and Meng, X.-L. Handbook of Markov Chain Monte Carlo. CRC Press, 2011.
• [2] Gao, C., Li, Q., and Guo, Z. Automobile insurance pricing with Bayesian general linear model. In Innovative Computing and Information. International Conference, ICCIC 2011, Wuhan, China, September 17-18, 2011. Proceedings, Part I (Berlin, 2011), M. Dai, Ed., Springer, pp. 359–365.
• [3] Gatsonis, C., Hodges, J. S., Kass, R. E., and Singpurwalla, N. D. Case Studies in Bayesian Statistics. Springer-Verlag, 1993.
• [4] Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B, Vehtari, A., and Rubin, D. B. Bayesian Data Analysis. 3rd Edition. Chapman & Hall/CRC. 2013.
• [5] Gelman, A., and Hill, J. Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press, 2007.
• [6] Goldburd, M., Khare, A., and Tevet, D. Generalized Linear Models for Insurance Rating. Casualty Actuarial Society, 2016.
• [7] Hoffman, M. D., and Gelman, A. The no-u-turn sampler: Adaptively setting path lengths in Hamiltonian Monte Carlo. Journal of Machine Learning Research 15, 1 (2011), 1593–1623.
• [8] Kruschke, J. K. Bayesian assessment of null values via parameter estimation and model comparison. Perspectives on Psychological Science 6, 3 (2011), 299–312.
• [9] Kruschke, J. K. Doing Bayesian Data Analysis: A Tutorial with R, JAGS, and Stan. Academic Press, 2015.
• [10] Kumar, R., Carroll, C., Hartikainen, A., and Martin, O. A. ArviZ a unified library for exploratory analysis of Bayesian models in Python. Journal of Open Source Software 4, 33 (2019), 1143.
• [11] Lynch, S. M. Bayesian theory, history, applications, and contemporary directions. In International Encyclopedia of the Social & Behavioral Sciences (Oxford, 2015), J. D. Wright, Ed., Elsevier, pp. 378–382.
• [12] McCullagh, P. Generalized Linear Models, Routledge, New York, NY, 1989.
• [13] Ntzoufras, I. Frontmatter. In Bayesian Modeling Using WinBUGS (Hoboken, HJ, 2009), I. Ntzoufras, Ed., John Wiley & Sons, pp. i–xxiii.
• [14] Ntzoufras, I. Introduction to Bayesian inference. In Bayesian Modeling Using WinBUGS (Hoboken, HJ, 2009), I. Ntzoufras, Ed., John Wiley & Sons, pp. 1–29.
• [15] Ohlsson, E., and Johansson, B. Non-Life insurance pricing. In Non-Life Insurance Pricing with Generalized Linear Models (Berlin, 2010), E. Ohlsson and B. Johansson, Eds., Springer, pp. 1–14.
• [16] Racine, A., Grieve, A. P., Flühler, H., and Smith, A. F. M. Bayesian methods in practice: Experiences in the pharmaceutical industry. Journal of the Royal Statistical Society. Series C (Applied Statistics) 35, 2 (1986), 93–150.
• [17] Salvatier, J., Wiecki, T. V., and Fonnesbeck, C. Probabilistic programming in Python using PyMC3. PeerJ Computer Science 2 (2016), e55.
• [18] Scheel, I., Ferkingstad, E., Frigessi, A., Haug, O., Hinnerichsen, M., and Meze-Hausken, E. A Bayesian hierarchical model with spatial variable selection: the effect of weather on insurance claims. Journal of the Royal Statistical Society Series C: Applied Statistics 62, 1 (2013), 85–100.
• [19] Tufvesson, O., Lindström, J., and Lindström, E. Spatial statistical modeling of insurance risk: a spatial epidemiological approach to car insurance. Scandinavian Actuarial Journal 2019, 6 (2019), 508–522.
• [20] Zhang, J., and Miljkovic, T. Ratemaking for a new territory: Enhancing GLM pricing model with a Bayesian analysis. Casualty Actuarial Society E-Forum 2 (2018), 1–32.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).