Optimal Coverage Limit Design for Health Insurance under Stochastic Claims and Policyholder Coverage Selection

Document Type : Research Article

Authors

1 Department of Actuarial Science, Faculty of Mathematical Sciences, Shahid Beheshti University, Evin, Tehran, Iran.

2 Department of Statistics & Actuarial Science, Stellenbosch University, Stellenbosch, South Africa.Stellenbosch, South Africa.

10.22054/jmmf.2026.92687.1287
Abstract
Determining optimal coverage limits is a fundamental problem in health insurance, as insurers must balance profitability, policyholders’ preferences, and risk exposure under uncertain claim experience. Existing studies generally optimize coverage limits by considering either claim uncertainty or policyholder behavior, while their combined effects have received limited attention. To address this gap, this paper develops a mathematical framework for the optimal design of multiple health insurance coverage limits by jointly incorporating stochastic claim frequencies and policyholders’ coverage selection probabilities. The insurer’s decision problem is formulated as the maximization of an exponential utility function subject to a probabilistic risk constraint defined by a loss threshold t and a confidence level θ. Using probabilistic bound theory, the stochastic optimization problem is transformed into an equivalent convex optimization problem that can be efficiently solved using the Lagrange multiplier method. The resulting optimality conditions lead to a system of nonlinear equations, for which the existence and uniqueness of the solution are established under appropriate regularity assumptions. The proposed framework is validated using health insurance data from Asmari Insurance Company for the period 2022–2023. The numerical results demonstrate that the proposed approach provides stable and computationally efficient optimal coverage limits while effectively controlling portfolio risk, illustrating its potential as a practical decision-support tool for health insurance contract design.

Keywords


[1] N. H. Agnew, R. A. Agnew, J. Rasmussen, and K. R. Smith, An application of chance constrained programming to portfolio selection in a casualty insurance firm, Management Science, (1969) vol. 15 pp. B512–B520.
[2] K. J. Arrow, Uncertainty and the welfare economics of medical care, In P. Diamond and M. Rothschild (eds.), Uncertainty in Economics, Academic Press, (1978) pp. 345–375.
[3] F. Baione, and D. Biancalana, An individual risk model for premium calculation based on quantile: a comparison between generalized linear models and quantile regression, North American Actuarial Journal, (2019) vol. 23 pp. 573–590.
[4] A. Bergesio, P. Koch-Medina, and C. Munari, Optimal insurance design under limited liability, Journal of Risk and Insurance, (2025) vol. 92 pp. 1122–1142.
[5] C. Bernard, X. He, J.-A. Yan, and X. Y. Zhou, Optimal insurance design under rankdependent expected utility, Mathematical Finance, (2015) vol. 25 pp. 154–186.
[6] Capco, US insurance survey 2023: Policyholders’ desire for personalized offerings, Capco Report, (2023) New York.
[7] C. Chen, K. Dong, E. W. Frees, R. Huang, S. Hui, and H. J. van Heerde, A joint model of cost and churn for the insurance industry, Journal of Risk and Insurance, (2026) Forthcoming.
[8] D. Crainich, Optimal self-insurance with genetic testing and state-dependent utility, Canadian Journal of Economics, (2025) vol. 58 pp. 418–442.
[9] H. Cremer, and J.-M. Lozachmeur, Nonlinear reimbursement rules for preventive and curative medical care, Journal of Health Economics, (2025) vol. 103 103049.
[10] C. Gollier, and H. Schlesinger, Arrow’s theorem on the optimality of deductibles: a stochastic dominance approach, Economic Theory, (1996) vol. 7 pp. 359–363.
[11] A. Y. Golubin, and V. N. Gridin, Optimal insurance strategy in a risk process under a safety level imposed on the increments of the process, Scandinavian Actuarial Journal, (2023) pp. 20–37.
[12] X. Han, and B. Li, Optimal insurance menu design under the expected-value premium principle, (2026), arXivpreprintarXiv:2604.15881.
[13] W. Hoeffding, Probability inequalities for sums of bounded random variables, Journal of the American Statistical Association, (1963) vol. 58 pp. 13–30.
[14] S.-C. Huang, J. Shi, and Y. Yang, A copula model for marked point process with a terminal event, The Annals of Applied Statistics, (2025) vol. 19 pp. 120–145.
[15] S. A. Klugman, H. H. Panjer, and G. E. Willmot, Loss Models: From Data to Decisions, Wiley Series in Probability and Statistics, 4th ed., (2012) Hoboken, NJ.
[16] G. Mahdavi, A mathematical model for deriving the optimal trajectory of life insurance demand, Journal of Mathematics and Modeling in Finance (JMMF), Allameh Tabataba’i University Press, (2025) vol. 5 no. 1 pp. 189-204.
[17] Y. J. Levy, and A. Veiga, Optimal contract regulation in selection markets, American Economic Journal: Microeconomics, (2025) vol. 17 pp. 94–126.
[18] S. S. Momahhed, S. E. Emamgholipour Sefiddashti, B. Minaei, and M. Arab, The optimal co-insurance rate for outpatient drug expenses of Iranian health insured based on the data mining method, International Journal for Equity in Health, (2024) vol. 25.
[19] S. D. Promislow, and V. R. Young, A unifying framework for optimal insurance, Insurance: Mathematics and Economics, (2005) vol. 36 pp. 347–364.
[20] R. Rahimisadegh, S. Noori Hekmat, M. H. Mehrolhassani, and M. Jafari Sirizi, Iran’s health insurance ecosystem: challenges and strategies, BMC Public Health, (2024) vol. 24 2470.
[21] A. Raviv, The design of an optimal insurance policy, In G. Dionne and S. E. Harrington (eds.), Foundations of Insurance Economics, Springer, (1979) pp. 251–263.
[22] Y. Zhang, and Y. Wu, Optimal health insurance and trade-off between health and wealth, Journal of Applied Mathematics, (2020) pp. 1–12.
[23] Y. Zhang, Y. Wu, and H. Yao, Optimal health insurance with constraints under utility of health, wealth and income, Journal of Industrial and Management Optimization, (2022) vol. 18 pp. 1875–1895.
[24] L. Zhou, A. Li, and J. Lu, Optimizing moral hazard management in health insurance through mathematical modeling of quasi-arbitrage, Risks, (2025) vol. 13 84.

Articles in Press, Accepted Manuscript
Available Online from 22 September 2026

  • Receive Date 11 May 2026
  • Revise Date 25 August 2026
  • Accept Date 12 September 2026