The Design Of Quantitative Futures Trading Strategy Based On Wave Theory And Neural Network

In recent years,with the enrichment of trading varieties,the perfection of trading system and the development of computer technology in China’s securities market,quantitative trading has gradually become an important means of investment in securities market.Quantitative trading with the help of modern finance,computer and mathematical methods,the investment concept and research results of people into an objective mathematical model,the use of computer technology to complete data processing,analysis and modeling,decision ordering,to achieve the whole transaction process systematized and programmed.Among many trading varieties,futures with standardized contracts,leverage trading based on margin,flexible two-way trading,standardized risk management system,attracted the attention of quantitative traders.Although the quantitative trading strategy for futures is more complex,it has good risk-return characteristics and has become the focus of the financial industry.In this paper,the design and optimization of quantitative futures trading strategy are mainly studied based on technical analysis and artificial intelligence.At first,when designing trading strategies,this paper sorts out elliott wave theory,quantitative trading,neural network and other related theories,based on elliott wave theory in technical analysis.Then,do the wave feature extraction of the preparation and the convolution of the neural network training work,in the convolution type neural network to catch the wave of technology,based on the convolution of the neural network method is used according to the wave model to predict futures prices rise and fall,and then build futures high-frequency trading strategies,and then use the trained model to predict the trend of the future,And using 300 stock index futures strategy back test and optimization.Finally,the empirical results are analyzed and summarized: the prediction trading model based on neural network and wave theory is more suitable for short-term trading of CSI 300 stock index futures.The prediction model based on convolutional neural network has a high success rate.The conclusion is helpful for investors to choose better strategies to trade in different periods and improve the return on investment.