数学理论与应用 ›› 2017, Vol. 37 ›› Issue (2): 112-121.

• • 上一篇    下一篇

基于SVM的沪深300股指期货量化交易策略

张剑, 王波   

  1. 上海理工大学管理学院,上海,200093
  • 出版日期:2017-06-30 发布日期:2020-09-24

Quantitative Trading Strategies of Shanghai and Shenzhen 300Index Futures Based on SVM

Zhang Jian, Wang Bo   

  1. Business School,University of Shanghai for Science and Technology,Shanghai 200093,China
  • Online:2017-06-30 Published:2020-09-24

摘要:

基于支持向量机(svm)理论建立沪深300股指期货量化交易模型,与传统对期货价格走势进行绝对预测的回归预测方法不同,模型利用支持向量机在处理非线性系统中的分类优势,将价格未来变化的趋势转化为交易信号,把一个复杂的时间序列回归预测问题转化为二分类问题.接着,把价量信息和技术指标分别作为输入向量,再引入止损机制,在动态预测模型上构建量化交易策略.采用历史数据对策略进行回测仿真,实证结果表明,价量信息交易策略表现要好于技术指标交易策略,量化交易模型总体取得了较好的盈利效果.

关键词: 机器学习, 支持向量机, 沪深300股指期货, 量化交易

Abstract: Based on the theory of support vector machine,aquantitative trading model of Shanghai and Shenzhen 300 stock index futures is established.Differing from the regression forecasting method,the model firstly makes use of the advantage of support vector machine in classification in nonlinear systems to transform a complex time series regression prediction problem into a two classification problem by converting the price evolution trend into a transaction signal,and then takes the price information and technical indicators as the input vector,introduces the stop-loss mechanism and obtains the quantitative trading strategy upon the dynamic forecasting model.Empirical results show that the price information transaction strategy has better performance than the technical index trading strategy,and overall,the quantitative trading model has achieved good profit effect.

Key words:

"> Machine learning, Support vector machine, Shanghai and Shenzhen 300stock index futures, Quantitative trading