Prediction of Monetary Fund Based on ARIMA Model

Abstract

Ant Financial has over 100 million subscribers, resulting in a significant amount of daily money influx and outflow. Therefore, capital management will be under a lot of stress. As a result, it's critical to precisely estimate inflows and outflows of funds while maintaining a low level of liquidity risk and keeping up with everyday business operations. Knowing when to make an accurate prediction is often the most important stage in forecasting. Instead, than duplicating what has already happened and will not happen again in the future, good forecasting highlights meaningful patterns and relationships in previous data. In this paper, we used ARIMA model to forecast total amount of purchase and redemption of Yu'E Bao in August 2014 and compared the predicted data with original data. After differential processing, the accuracy of the ARIMA model improves because the data becomes more stable and the rules become clearer.