Research on Forecasting Method of Balance Treasure Fund Flow

Capital inflow and outflow is very important for the survival of financial companies. This paper analyzes the data flowing out and out of some users of the balance. The use of more than 28,000 users of the balance in 14 months according to operating records, the user will be separated from large users, the remaining users by operating frequency is divided into inactive users, more active users, active users, ultra-active users. The ARIMA model is used to model the five classes respectively. Finally, the prediction results of the classification are obtained. There are significant differences in the inflow and outflow patterns of these five categories of people. The results of classification prediction are better than the overall forecast in the prediction of capital inflow.