Futures Rebar Price Forecast Based On Neural Network Model

The most important of the industrial raw materials is steel,and construction industry is the industry with heavy consumption of steel in China,especially the rebar,which is one of the largest steel varieties in China.In recent years,the futures market has been affected by factors such as supply and demand and strategic policies.Rebar prices have fluctuated dramatically and will soon face greater international market challenges.Therefore,a major issue for all participants in the construction industry in China is how to defense and avoid the market crisis caused by futures rebar price fluctuations.The paper puts forward the correct forecast of the price trend of futures rebar steel,whether there is a certain risk aversion and price warning function of the construction cost.Using BP neural network and wavelet neural network,the paper builds two kinds of futures rebar price forecasting models.One of the key points in the process is to use the Python programming language to quickly preprocess massive amounts of data.Three sets of experiments were performed on the hidden layer of each model.The error rate,convergence time,and accuracy of the experimental data were finally analyzed to determine the optimal two neural network models,and the final results were obtained through the unified input and output values.Finally,MIV value analysis and grey correlation analysis were used to analyze the importance of the influencing factors for the six attribute values in input values.Experiments show that the BP neural network obtains the data with high accuracy with the shortest running time.However,the running time of wavelet neural network is almost 10 times that of BP neural network,and the accuracy of the data obtained is also poorer.Based on the analysis of stability,accuracy and speed of convergence,the prediction model of BP neural network is better than that of wavelet neural network prediction model.Although the specific point of the price transition is difficult to determine,the overall trend of prices shows a stronger regularity.In terms of risk aversion,the forecasting model built in this paper can meet the needs of participants in the construction industry,such as owners,developers,constructors,suppliers,etc.,in forecasting rebar prices.