Application Of Combination Forecast Model In Crude Oil Futures Forecast

Crude oil is the bloodline of the world’s industrial economy.Since the reform and opening up,my country’s industry has developed rapidly,and the demand for oil is increasing,while the domestic supply is far less than the demand,and my country’s crude oil is increasingly dependent on imports.As a very important financial energy derivative,crude oil futures price changes will always affect my country’s economic development.Therefore,the study of crude oil futures price trends can provide reference for my country’s energy financial market to stabilize the economy and the sustainable development of the market economy.Starting from the midpoint and radius or information set settlement,the paper further builds a new model and expands the application scope of the model in the prediction of crude oil futures price range.Specifically,starting from the midpoint and radius of the interval number,and taking the minimum relative entropy as the optimal principle,a fixed-weight combined prediction model based on relative entropy is established.In order to further improve the prediction accuracy,the IOWHA operator is introduced,and a relative entropy-based Variable weight portfolio combined forecasting model.Through the crude oil futures data,it is verified that the combined forecasting model established in this chapter is more effective than the single forecasting model,and the optimal forecasting value of the crude oil futures data is given.Then,the COWA operator is used to convert the interval numbers into real numbers,and the sequence similarity index Theil’s inequality coefficient is used as the optimal criterion to construct a fixed-weight combined forecasting model.Afterwards,a variable-weight combined forecasting model was established by combining the GICOWA operator with Theil’s inequality coefficient.Then,from the perspective of error,the concept of interval number comprehensive error is defined,and the model is constructed based on the minimum sum of interval number comprehensive error under the norm.The IOWHA operator is introduced into the model,and a norm-based variable weight combined forecasting model is constructed.The empirical analysis of crude oil futures data shows that the two combined forecasting models based on norm can improve the prediction accuracy.After comprehensively considering the error index values of each model,the model with the best prediction effect is used to calculate the prediction value of the interval number of crude oil futures.Finally,at the end of this chapter,a brief comparison of the three combined forecasting models established in this paper is made.