Prediction model of insurance profit and loss under extreme weather based on PCA and ARIMA Models
DOI:
https://doi.org/10.54097/chmzvx33Keywords:
PCA, ARIMA, Catastrophe Risks, Prediction Model.Abstract
Because global catastrophe insurance gap widening, insurance profit forecast is important for ensuring the operation of insurance companies under extreme weather. In this paper, on the basis of analyzing the profit structure of insurance, the dimensionality reduction capability of PCA and the flexibility of ARIMA are combined to establish a prediction model of insurance profit and loss under the extreme weather. On the basis of using PCA (Principal Component Analysis) to extract the important features of extreme weather data and to conduct correlation analysis, the influences of each factor on the expression of different principal components are clarified, and the calculation formula of the principal component is obtained. ARIMA (Autoregressive Integrated Moving Average Model) is used to fit historical catastrophe insurance data and capture trends, to predict the likely future compensation ratio in this region. This method can not only reduce the dimension of complex weather data to simplify the analysis, but also flexibly predict the risk degree of future losses of insurance companies. The results show that with the increase of the frequency of extreme weather and the maximum rainfall, the operating risk of insurance companies will rise. Finally, in order to better develop regional catastrophe insurance, it’s a good choice to carry out long-term and short-term planning according to the prediction model.
Downloads
References
[1] Dai Yuxuan. A Preliminary Study on China's Diversified Catastrophe Insurance Model [J]. Journal of Disaster Prevention and Mitigation Engineering, 2024, 44(03):745-750.
[2] Zhang T, Ma X, Wang W J. Natural Disaster Insurance in China: Model Reference, Development Challenges and Suggestions[J]. International Journal of Natural Resource Ecology and Management, 2022, 7(2):93-98.
[3] Anita M, Snejana G, Michael M. Impact of Natural Disasters on the Value of (Re)Insurance Companies[J]. Zeitschrift für die gesamte Versicherungswissenschaft, 2023, 112(4):337-368.
[4] Patrick L, Brockett, Linda L et al. A comparison of neural network statistical methods and variable choice for life insurers financial distress prediction [J]. Journal of Risk and Insurance, 2006, 73(3):397-419.
[5] Hou Xuhua, Peng Juan. Research on Financial Risk Early Warning of the Internet Insurance Company Based on Entropy Method and Efficacy Coefficient Method [J]. The theory and practice of finance and economics, 2019, 40(5):7.
[6] Mikosch T. Heavy-tailed modelling in insurance [J]. Stochastic Models, 1997, 13(4):799-815
[7] Samuel R, Peter J.R, Wouter J.W.B. Insights into the complementarity of natural disaster insurance purchases and risk reduction behavior.[J].Risk analysis:an official publication of the Society for Risk Analysis, 2023, 44(1):141-154.
[8] Hou Xuhua, Xu Xiang. Research on the development evaluation and promotion path of green insurance for insurance companies under the goal of carbon peaking and carbon neutrality——Take Ping An Insurance Company of China for example[J]. Shanghai Insurance, 2023, (12):34-42.
[9] Sandow S, Zhou X. Data-efficient model building for financial applications: a semi-supervised learning approach[J]. The Journal of Risk Finance, 2007, 8(2):133—155.
[10] Huang Yifan, Meng Shengwang. The Design and Pricing of Earthquake Index Insurance in China [J]. statistical research, 2022, 39(04):108-121.
[11] Chao W .Valuing Multirisk Catastrophe Reinsurance Based on the Cox–Ingersoll–Ross (CIR) Model[J]. Discrete Dynamics in Nature and Society, 2021, 2021.
Downloads
Published
Conference Proceedings Volume
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.