基于技术交易系统性能分析中国股票市场有效性与研究.pdfVIP

基于技术交易系统性能分析的中国股票市场有效性研究 瞿慧 （研究领域：计算金融） 南京大学工程管理学院金融工程研究中心 摘要：考察技术交易规则相对于 “买入-持有”策略的超额收益是检验股票市场有效性的重 要方法。本文构建的技术交易系统能够根据市场环境自适应地调整输出，发挥多个技术交易 规则的优势，以避免传统方法要求一个技术交易规则持久地战胜市场才否定市场有效性的局 限性。系统以遗传编程为核心模块，消除对流行的技术交易规则先验知识的依赖，以避免数 据探测的问题。使用上证50 指数和深圳 100 指数历史价格与成交量的试验结果表明，该技 术交易系统输出的最优技术交易规则能够较“买入-持有”策略获得统计显著的样本外超额 收益，由此得出中国股票市场尚未达到弱式有效的结论。 关键词：市场有效性；技术交易系统；遗传编程 Using technical trading system to test the efficiency of Chinese stock markets QH Hui （Research Interest: Computational Finance ） Financial Engineering Research Center School of Management and Engineering, Nanjing University Abstract: Testing whether technical trading rules can beat buy and hold strategy is a common approach to study the efficiency of stock markets. We build a technical trading system which adjusts output optimal trading rules according to the market condition, so as to avoid the bias from only focusing on the performance of one particular technical trading rule . Instead of relying on the knowledge of popular technical trading rules, the system is base d on genetic programming, which avoids data snooping bias. Using daily price and volume data for SSE50 index and SZSE 100 index from January 2, 2004 to March 12, 2010, we find out that the technical trading system generated best technical trading rules have statistically significant out-of-sample excess return compared with buy-and-hold strategy considering realistic transaction cost. Therefore, we conclude that Chinese stock markets have not achieved weak-form efficiency. Key words: market efficiency; technical trading system; genetic programming 1 引言 技术交易规则通过分析价格、成交量等历史交易数据来指导交易决策，其战胜市场的能 力是对有效市场假说的直接否定，因此对市场有效性的实证研究常常基于对技术交易规则有 效性的检验。 现有文献常常通过实证某个或某些技术交易规则不能