Research on the Trading Strategy of Commodity Futures 

Shengtao Hu, Dai Hu* 

College of Science, Hunan City University, Yiyang, 413000, China

Abstract: Based on the data of rubber and thread steel, in this paper, we studied the commodity futures mar-ket. By MATLAB, the data of rubber was analysed, and the opening price, closing price, highest price and so on in different periods was calculated. And then according to these data, the sequence diagram of the price and other indicators of change over time was drawn by EXCEL. Three models of price prediction, i.e., varia-ble coefficient regression model, BP neural network model, time series model, are presented. And the fore-casts of the three models being compared, the forecasts are in good agreement with the actualities. Finally, the price of thread steel was forecasted by the moving average method, and the transaction of thread steel in a period of time was simulated under some simple trading rules.

Keywords: Variable coefficient regression model; BP neural network model; Time series model; Moving average method