MACHINE LEARNING-DRIVEN PREMIUM ESTIMATION IN CROP INSURANCE: LEVERAGING CLIMATE AND COMMODITY MARKET BIG DATA

Marija  Koprivica; Kristina Bradić; Mirjana  Gligorijević; Milica  Kočović De Santo

doi:10.59267/ekoPolj2603871K

Authors

Marija Koprivica Faculty of Economics and Business, University of Belgrade, Kamenička 6, 11000 Beograd https://orcid.org/0000-0003-4239-2252
Kristina Bradić Faculty of Economics and Business, University of Belgrade, Kamenička 6, 11000 Beograd https://orcid.org/0009-0008-9198-2770
Mirjana Gligorijević Faculty of Economics and Business, University of Belgrade, Kamenička 6, 11000 Beograd https://orcid.org/0000-0002-1144-442X
Milica Kočović De Santo Institute of Economic Sciences, Belgrade https://orcid.org/0000-0003-3304-7801

DOI:

https://doi.org/10.59267/ekoPolj2603871K

Keywords:

crop insurance, drought, machine learning, climate data, gradient boosting, logistic regression

Abstract

This study proposes a machine learning-driven frameworkfor premium estimation in crop insurance against droughtrisk, integrating climatic and commodity market data. Themethodology combines a Logistic Regression model forpredicting drought probability with an Extreme GradientBoosting model for forecasting commodity prices andcrop yields, using regional datasets from Vojvodina andŠumadija. The results show strong predictive performanceacross regions and metrics. Findings confirm that droughtoccurrence, yield and price volatility are driven by climaticand spatial heterogeneity, underscoring the need forregion-specific premium setting. The study demonstratesthe potential of machine learning techniques to supportmore risk-sensitive and actuarially consistent pricing inagricultural insurance.

Downloads

Download data is not yet available.

References

Aerts, J. C. J. H., & Botzen, W. J. W. (2011). Climate change impacts on pricing long-term flood insurance: A comprehensive study for the Netherlands. Global Environmental Change, 21(3), 1045–1060. https://doi.org/10.1016/j.gloenvcha.2011.04.005

Barratt, J., Kath, J., Collins, B., Mushtaq, S., & An-Vo, D. A. (2025). Decile-based index drought insurance to help improve income stability for wheat producers in Australia. Agricultural Water Management, 318, 109726. https://doi.org/10.1016/j.agwat.2025.109726

Birghila, C., Pflug, G. C., & Hochrainer-Stigler, S. (2022). Risk-layering and optimal insurance uptake under ambiguity: With an application to farmers exposed to drought risk in Austria. Risk Analysis, 42(12), 2639-2655, https://doi.org/10.1111/risa.13884

Botzen, W. J. W. (2016). Managing extreme climate change risks through insurance. Cambridge University Press.

Botzen, W. J. W. (2016). Trends in weather related disasters – Consequences for insurers and society. Weather and Climate Extremes, 11, 70–79. https://doi.org/10.1016/j.wace.2015.10.002

Brunette, M., Holecy, J., Sedliak, M., Tucek, J., & Hanewinkel, M. (2015). An actuarial model of forest insurance against multiple natural hazards in fir (Abies Alba Mill.) stands in Slovakia. Forest Policy and Economics, 55, 46–57. https://doi.org/10.1016/j.forpol.2015.03.001

Bucheli, J., Dalhaus, T., & Finger, R. (2021). The optimal drought index for designing weather index insurance. European Review of Agricultural Economics, 48(3), 573–597. https://doi.org/10.1093/erae/jbaa014

Bühlmann, H. (1985). Premium calculation from top down. ASTIN Bulletin, 15(2), 89–101. https://doi.org/10.2143/AST.15.2.2015021

Cabeza-Ramírez, L. J., Guerrero-Baena, M. D., Luque-Vílchez, M., & SánchezCañizares, S. M. (2024). Assessing farmers’ intention to adopt drought insurance. International Journal of Disaster Risk Reduction, 113, 104818. https://doi.org/10.1016/j.ijdrr.2024.104818

Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on KnowledgeDiscovery and Data Mining (pp. 785–794).

Crane-Droesch, A. (2019). Climate change projected to increase cost of the federal crop insurance program. Amber Waves. USDA Economic Research Service.

de Ruig, L. T., Haer, T., de Moel, H., Brody, S. D., Botzen, W. J. W., Czajkowski, J., & Aerts, J. C. J. H. (2022). How the USA can benefit from risk-based premiums combined with flood protection. Nature Climate Change, 12(11), 995–998. https://doi.org/10.1038/s41558-022-01501-7

Deaton, A., & Laroque, G. (1992). On the behaviour of commodity prices. Review of Economic Studies, 59(1), 1–23. https://doi.org/10.2307/2297923

Dubbelboer, J., Nikolic, I., Jenkins, K., & Hall, J. (2017). An agent-based model of flood risk and insurance. Journal of Artificial Societies and Social Simulation, 20(1), 6. https://www.jasss.org/20/1/6.html

Eze, E., Girma, A., Zenebe, A. A., & Zenebe, G. (2020). Feasible crop insurance indexes for drought risk management in Northern Ethiopia. International Journal of Disaster Risk Reduction, 47, 101544. https://doi.org/10.1016/j.ijdrr.2020.101544

Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232. https://www.jstor.org/stable/2699986

Führer, E., Horváth, L., Jagodics, A., Machon, A., & Szabados, I. (2011). Application of a new aridity index in Hungarian forestry practice. Időjárás, 115, 205–216.

Hohl, R., Jiang, Z., Tue Vu, M., Vijayaraghavan, S., & Liong, S. Y. (2021). Using a regional climate model to develop index-based drought insurance. Agricultural Finance Review, 81(1), 151–168. https://doi.org/10.1108/AFR-02-2020-0020

Hosmer, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (3rd ed.). Wiley.

Intergovernmental Panel on Climate Change (IPCC). (2023). Climate Change 2023: Synthesis Report. Cambridge University Press.

Islam, Md. M., Matsushita, S., Noguchi, R., & Ahamed, T. (2022). A damagebased crop insurance system for flash flooding. Asia-Pacific Journal of Regional Science, 6(1), 47–89. https://doi.org/10.1007/s41685-021-00220-9

Kapphan, I., Calanca, P., & Holzkämper, A. (2012). Climate change, weather insurance design and hedging effectiveness. The Geneva Papers on Risk and Insurance, 37(2), 286–317. https://doi.org/10.1057/gpp.2012.8

Koprivica, M., Kočović, J., & Cvijanović, D. (2024). A model for determining premium rates in index-based crop insurance. Ekonomika poljoprivrede, 71(2), 397–412. https://doi.org/10.59267/ekoPolj2402397K

Madaki, M. Y., Kaechele, H., & Bavorová, M. (2023). Agricultural insurance as a climate risk adaptation strategy in developing countries. Climate Policy, 23(6), 747–762. https://doi.org/10.1080/14693062.2023.2220672

Pindyck, R. S., & Rotemberg, J. J. (1990). The excess co-movement of commodity prices. The Economic Journal, 100(403), 1173–1189. https://doi.org/10.2307/2233966

Radosavljević, K., Bradić, K., & Đurđić, D. (2025). Sustainable agriculture in the digital age: Innovations in insurance. In J. Kočović et al. (Eds.), Innovations in insurance – from traditional to modern market (pp. 75–92). University of Belgrade.

Stojanović, Ž., Gligorijević, M., & Rakonjac-Antić, T. (2012). The role of the marketing mix in the improvement of agricultural insurance. Ekonomika poljoprivrede, 59(4), 769–780. https://www.ea.bg.ac.rs/index.php/EA/article/view/545

Tesselaar, M., Botzen, W. J. W., Haer, T., Hudson, P., Tiggeloven, T., & Aerts, J. C. J. H. (2020). Regional inequalities in flood insurance affordability and uptake under climate change. Sustainability, 12(20), 8734. https://doi.org/10.3390/su12208734

Vinnarasi, R., Dhanya, C. T., & Kumar, H. (2023). Tracing time-varying characteristics of meteorological drought through nonstationary joint deficit index. Journal of Climate, 36(12), 4203–4217. https://doi.org/10.1175/JCLI-D-22-0437.1

Wu, R., & Yang, Y. (2024). Drought disaster risk assessment and yield insurance pricing of maize. Geomatics, Natural Hazards and Risk, 15(1), 2420660. https://doi.org/10.1080/19475705.2024.2420660