<![CDATA[One consequence of the rising number of traffic accidents is the increase in the burden of insurance claims on insurance companies, especially social insurance for traffic accidents. Unexpected fluctuations in claim numbers create challenges for companies, so forecasting is needed to identify future trends in claim numbers. The number of claims submitted is recorded periodically, thereby forming time series data. This allows for the application of time series analysis, such as the Autoregressive Integrated Moving Average (ARIMA) method. This study aims to to model and forecast the number of traffic accident social insurance claims using the ARIMA method. This study utilizes recapitulation data on traffic accident social insurance claim submissions spanning 120 weeks from July 2023 – December 2025. The ARIMA method is employed to identify the optimal ARIMA (p,d,q) model for the data. Analysis results indicate that ARIMA (3,1,0) is the optimal model, with a Mean Squared Error (MSE) of 150,054 and a Mean Absolute Percentage Error (MAPE) of 10,53%. Forecasting results for the next 20 weeks reveals a stable tren in claims. This suggest that there is no significant increase or decrease in insurance claims during the forecast period.]]>
Copyrights © 2026
