Research on the Development of Insurance Industry Based on ARIMA Time Series Model

In recent years, the probability of extreme weather events occurring globally has been increasing, posing a threat to human life and property and presenting significant challenges to the insurance industry. Insurance serves as an important pillar of societal stability by reducing the risk borne by policyholders, therefore achieving sustainable development in the insurance sector is an urgent issue to address. This article focuses on studying insurance models and protection measures for insurance companies and regions, an analysis of the time series of loss amounts caused by extreme weather reveals an irregular pattern. Therefore, an ARIMA time series model is employed for prediction, and a random forest model is used to fit the insurance claim amounts. The model is validated with a coefficient of determination of 0.954745, indicating a good fit. By consid-ering premium income and insurance claim amounts, a profitability calculation formula is derived, which can further determine whether an insurance company should provide coverage in a specific region. Using entropy weight method to determine the weights of various attributes in multi-attribute decision-making process. Then calculate the risk score for each region and divide the risk score into three categories: low, medium, and high. A three-dimensional scatter plot of the amount, longitude, and latitude of property damage was constructed to more accurately determine whether construction should be carried out in each area. Finally, based on this analysis process, reasonable construction strategies were proposed.