在中国A股市场基于舆情的交易策略

发布时间：2018-03-17 18:11

本文选题：中国股票市场　切入点：机器学习　出处：《上海交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着互联网和移动互联网时代的到来，数据变得越来越重要，“大数据”这个概念也正在改变我们的生活以及很多行业的生产经营行为。互联网让数据的共享便利化，任何人都能在互联网的海量数据中找到所需要的信息，且人与人的沟通也随着Web2.0的出现而更加频繁和密切，所以诞生了许多舆论情报分析的职业和岗位。在中国股票市场上，散户投资者居多，且多为非理性投资者，其投资决策更容易受到消息面的影响，他们的存在增加了股市的波动性。散户投资者往往在各大股吧表现活跃，他们的情绪也直接影响着股市的成交状况。研究他们的活动可以加深我们对中国资本市场的理解。基于行为金融学和数据挖掘原理，本文用机器学习的方法，通过分析这些投资者的发帖、评论等文本信息，，从中挖掘出他们的情绪指标及其变化。然后通过多元回归，证明这些情绪指标能够为后一交易日的收盘价提供额外的显著信息。并且，在考虑交易成本之前和之后，本文提供的交易策略在243个交易日中相对大盘分别超过27.49%和14.62%。
[Abstract]:With the advent of the Internet and the mobile Internet, data is becoming more and more important. The concept of "big data" is also changing our lives and the production and operation of many industries. The Internet makes data sharing easier. Anyone can find the information needed in the massive data of Internet, and the communication between people and people is more frequent and closer with the appearance of Web2.0, so many professions and posts of public opinion intelligence analysis have been born. In China's stock market, retail investors are more likely to be irrational, and their investment decisions are more susceptible to news. Their presence increases the volatility of the stock market. Retail investors tend to be active in big stock bars. Their emotions also have a direct impact on trading in the stock market. Studying their activities can deepen our understanding of China's capital markets. Based on the principles of behavioral finance and data mining, this paper uses the method of machine learning, by analyzing the text information of these investors, such as posts, comments and other text information, excavates their emotional indicators and their changes from them, and then through multiple regression. It is proved that these sentiment indicators can provide significant additional information for the closing price of the next trading day. Moreover, before and after considering transaction costs, the trading strategies provided in this paper exceed 27.49% and 14.62 respectively in 243 trading days.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【共引文献】