Empirical Research On Price Forecasting And Quantitative Trading Strategies Of Copper Futures

As the copper spot price "Chinese vane" Shanghai copper futures price volatility,it brings in uncertainty to stakeholders,so we need to make further research on Shanghai copper futures price operation rules and trading strategy.The difficulty and complexity of financial product price prediction has exceeded the scope of traditional econometric models.Multi model combination forecasting method can integrate the advantages of each model and make the prediction accuracy higher.At the same time,the rapid expansion of the futures market provides a soil for quantitative trading,and quantitative trading will be an important direction for future research.This paper uses GARCH model,SVM model and the combined model of Shanghai copper futures price dynamic prediction model,and then use the initial construction of Shanghai copper futures trading strategy support vector machine model for the construction of the futures market and quantitative timing strategy construction to provide some reference.The empirical results show that,in the dynamic prediction of GARCH model and PSO-SVM model,MAPE is 3.92 and 2.89,respectively.In the prediction of GARCH-PSO-SVM combination model,MAPE is reduced to 2.68,and the performance of combined model is better than that of single model.In addition,in the construction of comparative price and volume information and technical indicators of the performance of information quantitative trading strategies in Shanghai copper futures.Through the back test of the transaction strategy,it is found that the transaction strategy based on the basic price information is better than the trading strategy based on the technical index.The basic price amount of information trading strategy based on the total rate of return is 90.992%,the maximum retracement is only 5.36%,while the technical indicators of the total rate of return based on the trading strategy is 73.586%.the maximum retracement is 17.19%,It shows that the transaction strategy of basic price information is higher and more stable,because the lack of technical indicators is sensitive to price changes.