Classifying Age of Policyholders According to the Claim Rates in Iran

Document Type : Research Article


1 Data Mining Desk leader, Electrical Insurance Research Group, Insurance Research Center, Iran

2 . Data Mining Cosultant, Electrical Insurance Research Group, Insurance Research Center, Iran

3 Associate Professor, Faculty of Personal Insurance Research Group, Insurance Research Center, Iran


According to Note 1, Article 18 of Third Party Liability (TPL) Insurance Law 1395 (2016), Central Insurance of I. R. Iran (CII) in cooperation with Law Enforcement Force of I. R. Iran is obliged to work out to allow for the issuance of TPL policy based on the characteristics of the driver by the end of Sixth Five-Year Development Plan. As a result, in the premium calculation and the third-party claims, the characteristics of the driver must be taken into account. Thus, using six-year data from an insurance company, the effect of the policyholders’ age on the incurred damages are investigated. In this study, initially, with the help of Kolmogorov-Smirnov two-sample test, the age distribution of the reckless policyholders and drivers were compared. The results indicated that ‘age’ characteristic plays a role in causing damages. Due to the age differences among policyholders and the four major vehicle types, the calculations were distinctively performed based on each set of vehicles. Using conditional probability, the possibility of damages caused by different age groups was measured. Furthermore, the decision tree model was developed based on the three registered characteristics of the policyholders including age, gender, and type of customer. The results indicated that among the passenger cars, trucks, and motorcycles, the probability of damages caused by policyholders having less than 22, 30, and 25 years of age is respectively considerably high. Thus, according to actuarial principles, these policyholders must reasonably pay more premiums. The results further showed that more than 50% of the claims that involved fatality in all types of vehicles were related to drivers that had no insurance policy. Therefore, several drivers that share the same vehicle must pay more premiums. The age characteristic of the policyholders does not have any significant effect on claims involving buses or vans.


اصغری اسکوئی، م. ر.؛ خانی‌زاده، ف.؛ و بهادر، آ.، (1399). «کاربرد داده کاوی با استفاده از الگوریتم‌های یادگیری ماشین برای بررسی تأثیر ویژگی‌های خودرو در پیش‌بینی ریسک خسارت مالی در رشته بیمه شخص ثالث». پژوهشنامه بیمه، شمارۀ 35، صص: 33-66.
عربیان، ن.؛ خداپرست مشهدی، م.؛ همایونفر، م.؛ فیضی، م.، و فکور، و.، (1399). «آزمون انتخاب بد در بازار بیمه پایه درمان ایران». فصل نامۀ علمی مطالعات اقتصادی کاربردی ایران، شمارۀ 33، صص: 93-111.
کریم‌زادگان مقدم، د.؛ و بهروان، م.، (1394). «ارائۀ راهکاری برای تعرفه گذاری پویا در صنعت بیمه با استفاده از تکنیک داده کاوی (مطالع موردی : بیمه شخص ثالث)». پژوهشنامۀ بیمه. شمارۀ 30، صص: 1-20.
محقق‌زاده، ف.؛ شیرین‌بخش، ش. ا.؛ نجفی‌زاده، ع.؛ و دقیقی‌اصلی، ع.، (1396). «تأثیرپذیری رشد اقتصادی از بیمه‌های زندگی و غیرزندگی». فصلنامه مطالعات اقتصادی کاربردی ایران، شمارۀ 23، صص: 205 -230.
Asghari Oskoei, M. R.; Khanizadeh, F. & Bahador, A., (2020). “Application of Data Mining through Machine Learning Algorithms to Study Effect Insurance”. Iranian Journal of Insurance Research. No. 35(1), Pp: 33-66. doi: 10.22056/jir.2020.228093.2724.
Arabian, N.; Khodaparast Mashhadi, M., Homayounifar, M.; Feizi, M. & Fakoor, V., (2020). “Adverse selection Test in Iran's Basic Health Insurance Market”. Journal of Applied Economics Studies in Iran, 2020, No. 9(33), Pp: 93-111. doi: 10.22084/aes.2019.17763.2768.
Bedsworth, F.; Portillo, J. E.; Neal, D. R. & Willardsen, K., (2021). “Asymmetric information and insurance: An experimental approach”. Journal of Behavioral and Experimental Econom.    
Karim Zadehgan Moghaddam, D., Behrvan, M., (2016). “Dynamic Tariffing in Insurance Industry With Data Mining Method (Case Study: Third Person Car Insurance)”. Iranian Journal of Insurance Research, 2016, No.  30(120), Pp: 1-20. doi: 10.22056/jir.2016.17618.
Mohagheghzadeh, F.; Shirinbakhsh, S.; Najafizadeh, A. & Daghighi asl, A., (2017). “Sensitivity of economic growth to life and non-life insurances”. Journal of Applied Economics Studies in Iran, 2017, No. 6(23), Pp: 205-230. doi: 10.22084/aes.2017.13128.2415.
Shams Esfandabadi, Z.; Ranjbari, M. & Domenico Scagnelli, S., (2020). Prioritizing Risk Level Factors in Comprehensive Automobile Insurance Management: A Hybrid Multi-Criteria Decision-Making Model. Global Business Review.
Śliwiński, A. & Kuryłowicz, Ł., (2021). Usagebased insurance and its acceptance: An empirical approach. Risk management and Insurance Review.
Anon., (2018). Young car driver’s road safety factsheet, s. l.: Department for transportation.
Bedsworth, F.; Portillo, J. E.; Neal, D. R. & Willardsen, K., (2021). “Asymmetric information and insurance: An experimental approach”. Journal of Behavioral and Experimental Economics, Vol. 92, Article 101696.
Charani, N. S., (2020). “Bayesian Smoothing of Decision Tree Soft Predictions and Evidential Evaluation”. Cham, Springer, Pp: 368-381.
Denuit, M.; Hainaut, D. & Trufin, J., (2019). “Insurance Risk Classification”. Effective Statistical Learning Methods for Actuaries, Pp: 3-26.
Dussault, C. & Letendre, P., (1999). “Automobile Insurance: Road Safety, New Drivers, Risks”. Insurance Fraud and Regulation. s.l.: Springer.
Esfandabadi, Z. S. & Esfahani, M. M. S., (2018). “Identifying and classifying the factors affecting risk in automobile hull insurance in Iran using fuzzy Delphi method and factor analysis”. Journal of Industrial Engineering and Management Studies, Vol. 5 (2). Pp. 84-96.
Kozak, J., (2017). Decision Tree and ensemble learning based on Ant colony optimization. s.l.: Springer.
Maag, U. et al., (1999). An evaluation of the effects on crashes of the 1991 legislative reform on new licensees in quebec. New York: Springer Science.
Parodi, P., (2014). Pricing in general insurance. s.l.: CRC press.
Pratt, John, W. & Gibbons, Jean D., (1981). “Kolmogorov-Smirnov Two-Sample Tests”. In: Concepts of Nonparametric Theory. New York: Springer, Pp: 318-344.
Thomas, G. R., (2009). “Demand elasticity, Risk classification and loss coverage: when can community rating work”. Astin bulletin, Pp: 403-428.
Williams, A. F., (1999). Licensing policies for young drivers in the united states. NewYork: Springer.