Research on the Improved Neural Network of Coal Price Forecast Based on Co-integration Theory

Abstract

China is a big coal consumer, and changes in coal prices will have a huge impact on the entire market economy system. In this paper, aiming at the prediction of coal price, the grey correlation analysis method is firstly used to describe the correlation degree between the main factors and the coal price, and the main influencing factors are sorted according to the correlation coefficient. Finally, it is determined that a coal price prediction system needs to be established on such indicators as coal reserves, raw coal production, total coal consumption, the proportion of coal in the energy mix, coal road traffic volume, coal imports, etc. Next, the quantitative indicators are combined with the economic cycle and cointegration theory to study, with the help of cointegration theory to improve the traditional neural network, so that it is more suitable for the prediction of coal prices. Finally, this paper obtains an effective coal price prediction model, which can better predict and analyze the future coal price and can play an important reference role in the formulation of government policies.