DOI QR코드

DOI QR Code

Estimating home fire severity with statistical distributions

통계적 분포를 통한 주택 화재 심도 추정

  • Yunjung Park (Department of Statistics, Ewha Womans University) ;
  • Inha Song (Department of Statistics, Ewha Womans University) ;
  • Soyoun Lee (Department of Statistics, Ewha Womans University) ;
  • Kwang Hyun Nam (Underwriter Inc.) ;
  • Rosy Oh (Korea Military Academy) ;
  • Jaeyoun Ahn (Department of Statistics, Ewha Womans University)
  • 박윤정 (이화여자대학교 통계학과) ;
  • 송인아 (이화여자대학교 통계학과) ;
  • 이소연 (이화여자대학교 통계학과) ;
  • 남광현 (주식회사 언더라이터) ;
  • 오로지 (육군사관학교 수학과) ;
  • 안재윤 (이화여자대학교 통계학과)
  • Received : 2023.05.01
  • Accepted : 2023.07.05
  • Published : 2023.12.31

Abstract

This paper evaluates the performance of various distribution assumptions in regression settings for estimating insurance loss. The gamma distribution is commonly used to handle the asymmetry property of loss distribution. However, recent studies highlight the significance of heavy-tailedness in loss distribution. Through an analysis of real home fire insurance data, we compare the effectiveness of different distribution assumptions in regression methods. Our findings show that the choice of parametric distributional assumption is crucial in determining premiums for various insurance products, including "excess of loss insurance" and "limit insurance". Additionally, we discuss practical considerations for applying our results in home fire insurance.

본 논문은 보험 손실 추정을 위한 회귀 설정에서 다양한 분포 가정의 성능을 실제 데이터를 사용하여 비교 분석합니다. 감마 분포는 일반적으로 보험의 손실 분포의 비대칭성을 처리하는 데 사용됩니다. 그러나 최근 연구는 보험자료의 분석에 있어서 손실 분포의 두꺼운 꼬리의 중요성을 강조합니다. 실제 주택 화재 보험 데이터 분석을 통해 우리는 회귀 방법에서 다양한 분포 가정의 효과를 비교합니다. 우리의 결과는 보험손실에 대한 분포 가정의 선택이 "초과 손해 보험" 및 "한도 보험"을 포함한 다양한 보험 상품의 보험료 결정에 중요하다는 것을 보여줍니다. 또한 주택 화재 보험의 통계적 모형설정에 있어서 실제 고려 사항에 대해 논의합니다.

Keywords

Acknowledgement

이 성과는 서울대학교 경영대학 경영연구소 연구비 지원과 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원 (No. RS-2023-00217022) 을 받아 수행된 연구임.

References

  1. Ahn S, Kim JH, and Ramaswami V (2012). A new class of models for heavy tailed distributions in finance and insurance risk, Insurance: Mathematics and Economics, 51, 43-52.  https://doi.org/10.1016/j.insmatheco.2012.02.002
  2. Agresti A (2015). Foundations of Linear and Generalized Linear Models, John Wiley & Sons, New Jersey.
  3. Asmussen SR (2003). Applied probability and Queues, Stochastic Modelling and Applied Probability, 51, 266-301.  https://doi.org/10.1007/0-387-21525-5_10
  4. Foss S, Korshunov D, and Zachary S (2013). An Introduction to Heavy-Tailed and Subexponential Distributions, Springer Science & Business Media, New York. 
  5. Gray RJ and Pitts SM (2012). Risk Modelling in General Insurance: From Principles to Practice, Cambridge University Press, New York. 
  6. Klugman SA, Panjer HH, and Willmot GE (2012). Loss Models: From Data to Decisions, John Wiley & Sons, New York. 
  7. Lee A, Ahn J, Mun H, Nam K, Park Y, Song I, and Park S (in press). Introducing new rate factors and statistical learning to improve fire loss prediction accuracy, The Korean Journal of Applied Statistics. 
  8. Little RJA (1988). Missing-data adjustments in large surveys (with discussion), Journal of Business Economics and Statistics, 6, 287-301.  https://doi.org/10.2307/1391878
  9. Nelder JA and Wedderburn RW (1972). Generalized linear models, Journal of the Royal Statistical Society: Series A (General), 135, 370-384.  https://doi.org/10.2307/2344614
  10. National Fire Service Academy (2019). Fire investigation practice, Technical Paper, Retreived Mar. 10, 2023, Available from: https://www.nfsa.go.kr/nfsa/releaseinformation/archive/materials/ 
  11. Oh R, Jeong H, Ahn JY, and Valdez EA (2021). A multi-year microlevel collective risk model,Insurance: Mathematics and Economics, 100, 309-328.  https://doi.org/10.1016/j.insmatheco.2021.06.006
  12. Punzo A, Bagnato L, and Maruotti A (2018). Compound unimodal distributions for insurance losses, Insurance: Mathematics and Economics, 81, 95-107.  https://doi.org/10.1016/j.insmatheco.2017.10.007
  13. Park H (2022). Case studies on the fire safety in attic space of multi-family housing (Master's thesis), Seoul National University of Science and Technology, Seoul. 
  14. Rubin DB (1986). Statistical matching using file concatenation with adjusted weights and multiple imputations, Journal of Business Economics and Statistics, 4, 87-94.  https://doi.org/10.1080/07350015.1986.10509497
  15. Sarabia JM, Jorda V, Prieto F, and Guillen M (2020). Multivariate classes of GB2 distributions with applications, Mathematics 2021, 9, 72. 
  16. Sigman K (1999). Appendix: A primer on heavy-tailed distributions, Queueing Systems, 33, 261-275.  https://doi.org/10.1023/A:1019180230133