Author(s)

Jiayi Ma

Corresponding Author

Jiayi Ma

Abstract

This study aims to address the risk challenges brought by increasingly frequent extreme weather events and to build a comprehensive assessment model to provide scientific support for insurance companies, real estate developers and community managers in disaster prevention and risk decision-making. The study is divided into two core tasks. In Task 1, an extreme weather risk assessment model is constructed by combining the entropy weight method and time series analysis (EWM-ARIMA) to quantitatively predict the intensity and frequency of extreme weather events in different regions. At the same time, the Premius-Policyholder risk assessment framework is introduced, and the analytic hierarchy process (AHP) is used to integrate economic, social and sustainability risks to form a decision support model for insurance companies to optimize their underwriting strategies. Task 2 is based on principal component analysis (PCA) and decision tree algorithm to establish a real estate construction risk assessment model, and combine house building types and geographical zoning to complete model integration. Finally, the stability and robustness of the model under key parameter fluctuations are verified through sensitivity testing. The results show that the errors are all controlled within 5%, indicating that the model has good practicality and reliability.

Keywords

Extreme-Weather Events, Evaluation Model, Premium Setting, EWM-ARIMA, Principal Component Analysis