The best way to conference proceedings by Francis Academic Press

Web of Proceedings - Francis Academic Press
Web of Proceedings - Francis Academic Press

Analysis of Influencing Factors of Import and Export Based on Neural Network Gray Prediction

Download as PDF

DOI: 10.25236/iemb.2021.004


Weiyi Yang, Yanru Liu, Yifei Xin

Corresponding Author

Weiyi Yang


Import trade and domestic consumption may be closely related to the industrial structure. The purpose of this paper is to study the impact of RMB exchange rate jump risk on China's import and export trade from a macro level. Based on GARCH-JUMP model, this paper proposes a combined forecasting method of energy demand based on grey correlation analysis and BP neural network to solve the problems of single influencing factors, neglect of macro-policy factors and lack of forecasting accuracy in energy demand forecasting. Impulse response analysis and variance decomposition analysis based on VAR model are carried out. The empirical results show that the impact of RMB exchange rate jump risk on import and export trade is more significant in the short term than in the long term, and the impact on import and export trade changes from positive to negative and then to positive. And the impact on import trade is greater than that on export trade. The domain is divided according to the pertinence and direction of the macro policy, and the indicators of each field are analyzed; secondly, the grey correlation analysis method is used to calculate, sort and screen the initial indicators of the macro policy in each field after analysis; finally, the selected initial index is used as the input of BP neural network, and the neural network is used to predict energy demand, and an example is carried out to simulate and analyze the forecasting results of energy demand combination forecasting. The result of an example shows that the proposed energy demand combination forecasting method focuses on the impact of macro policy, and effectively improves the prediction accuracy. It is practical and reliable to conduct empirical research and analysis on the impact of RMB real effective exchange rate on China's import and export, in order to better stabilize China's RMB real effective exchange rate, reduce the impact on China's import and export trade, and stabilize China's economic development. We will promote the better improvement of China's balance of payments environment and promote the better development of China's import and export trade.


Import and export trade, international commerce, BP neural network, forecasting model, exchange rate