Author(s)

Zhiyang Wang, Hailong Chen, Haodong Hong, Yi Zhang, Yuxin Sun, Xiang Zhou, Yuqi Wu

Corresponding Author

Zhiyang Wang

ABSTRACT

The money supply is an indicator that reflects the country’s economic and financial conditions. Changes in the money supply directly affect the country’s economic development and people’s lives. Therefore, studying money supply has very important practical value. The money supply mainly depends on the loan balance of the financial institution, but the loan balance of the financial institution and the money supply present a more complicated non-linear relationship, and it is difficult to obtain the specific relationship between them by conventional methods. Based on this, this paper optimizes the weights and thresholds of the BP neural network through the ant colony algorithm, obtains the optimal weights and thresholds of the BP neural network prediction model, and uses the optimal weights and thresholds for the BP neural network training and prediction. The simulation results show that the convergence speed and prediction accuracy of the optimized model have been greatly improved. The money supply can be accurately predicted by this forecasting method, which has very important guiding significance for the rapid development of social economy and the stability of people’s lives.

KEYWORDS

Ant colony algorithm, BP neural network, money supply, loan balance of financial institutions, prediction model