The Quantitative Timing-choose Strategy Based On Support Vector Machine And Its Empirical Research

The artificial intelligence storm of Google AlphaGo is sweeping across the industry and it will also have a profound impact on the financial investment industry.Quantitative investment has been paid more and more attention by the academic circle and the investment community because of its rational and objective decision,high efficiency and strong ability of information processing.Quantitative timing-choose strategy is an important branch of quantitative investment strategy.Support vector machine(SVM)is a kind of machine learning algorithm,which made up for the shortcomings of the traditional neural network learning algorithm.The researches of SVM in domestic financial field were mainly used in financial time series forecasting and there was no study combined with quantitative timing-choose strategy.SVM in the process of researches was mainly focused on the SVM method and its application.But the researches often ignored the strategy itself.In view of the above problems,this paper studied the existing quantitative timing strategy and SVM algorithm,combined with the advantages of them and built the quantitative timing-choose strategy based on SVM.First of all,this paper introduced the related concepts of quantitative investment and briefly reviewed the development of quantitative investment at home and abroad.This paper also defined the quantitative timing-choose strategy,analyzed its characteristics and classified the existing quantitative timing-choose strategy.Secondly,this paper made a comprehensive and in-depth study of the relevant theories of SVM from six aspects.Next,this paper constructed the quantitative timing-choose strategy based on SVM,which was mainly composed of two parts,one was the construction of SVM timing-choose strategy and the other was the setting of the model algorithm.Finally,this paper proved the effectiveness of the strategy through the date of Chinese oil,Shanghai Pudong Development Bank,Shanghai and Shenzhen 300 index,CSI 500 index and Gem index.The innovative work of this paper mainly has two aspects:First of all,this paper made a systematic review of quantitative timing-choose strategy and established a quantitative timing-choose strategy.The overall idea of the timing-choose strategy in this paper is: The operation of the strategy is in China's stock market.SVM prediction model runs once at the end of each day and predicts the next day's closing price.If the prediction result is rising,when the price is lower than the closing price of the previous day in the next day,we buy the whole store and vice versa.If the prediction result is that the price is constant,we will not operate.The strategy adds a stop loss judgment at the same time.The transaction is performed once at most in a day.The whole process is automatically carried out by the trading system.Secondly,this paper introduced the SVM optimization algorithm to construct and test the quantitative timing-choose strategy.In the part of the construction of the strategy,the author constructed the strategy which included seven aspects: the general idea of the timing-choose model design,the forecast period,the forecast target,the investment scope,the characteristic index,the buying and selling point and the model setting.In the part of the setting for the model algorithm,this paper included five aspects: the selection of multi classification algorithm of SVM,the selection of kernel function of SVM,the optimization of parameters,the processing of unbalanced data and the rolling prediction.Based on the analysis of the predictive ability of SVM,the contrast with the buy and hold strategy,the performance under different market conditions and the evaluation indicators of strategy,the performance of the strategy was excellent and the strategy was effective.The research of this paper will be helpful for the construction of the quantitative investment strategy.It also has some guidance and reference value for the improvement and optimization of quantitative timing-choose strategy and quantitative investment practice.