上海股票市场的有效性与排列熵研究

发布时间：2018-02-16 04:29

本文关键词： 上海股票市场复杂性有效性排列熵　出处：《南京财经大学》2013年硕士论文　论文类型：学位论文

【摘要】：现代金融学是在有效市场理论的基础上建立起来的，在此基础上也产生和发展了很多重要的金融模型。但是，金融市场因为“有限理性”投资者的参与而成为一个复杂适应性系统。复杂性的存在使得人们认识和驾驭金融市场的难度增加，然而，复杂性又支撑着金融系统的相对稳定性。因而，考虑金融市场的复杂性和有效性两者之间是否存在交互的影响，以及如果存在影响时又是以何种形式存在的成为一个值得研究的课题和内容。熵开始是一个热力学的概念，用于描述热力学系统的混乱程度。在经济学领域中，熵表示复杂系统的均匀和不确定程度，因此它可以作为度量系统复杂性程度的指标。另一方面，针对有效市场假说的不足而提出的分形市场假说近年来在金融市场的有效性分析层面起到了重要的作用。因而，本文以熵和有效性指数分别作为市场复杂性和有效性的度量，以上证综合指数1991年7月15日至2012年6月29日的收盘价数据为研究对象，根据上证综指变化的特点将其划分为7个时间区间，就以下几个方面对上述问题进行了研究：首先，基于赫斯特指数与0.5的绝对离差构造有效性指数，来代表市场有效性，并利用多重分形消除趋势波动分析（MF-DFA）实证研究了上海股票市场的有效性，结果表明，上海股票市场目前远远没有达到“弱式”有效状态。其次，应用排列熵理论探讨了上海股票市场复杂性。实证分析显示排列熵能够及时快速有效的提取各种股价变化时期的特征，体现股价变化的有序性，从而能够正确的反应股票市场的复杂性程度。最后，结合“滑动窗口”技术，分析了上海股票市场的效率和排列熵之间的相关关系。应用BDS检验发现，效率指数和排列熵之间存在非线性关系；使用交叉相关系数分析表明，两者之间存在交叉影响，且长期的影响较短期更显著；利用Diks等提出的非线性Granger检验方法指出，两者之间存在非常稳健地双向格兰杰因果关系。因此，短期市场复杂性的增加可能会降低市场效率，但是长远来看，可以通过增加市场复杂度的方法来提升股票市场的效率。本文也从实证的角度证实了刘维奇的观点，即市场效率与市场本身的复杂性密切相关。
[Abstract]:Modern finance is established on the basis of efficient market theory, on which many important financial models have emerged and developed. Financial market has become a complex adaptive system because of the participation of "limited rational" investors. The existence of complexity makes it more difficult for people to understand and control financial market. Complexity, in turn, supports the relative stability of the financial system. Therefore, considering whether there is interaction between the complexity and effectiveness of financial markets, And if there is influence in what form of existence is worth studying the subject and content. Entropy began as a thermodynamic concept used to describe the degree of confusion in thermodynamic systems. In the field of economics, entropy represents the degree of uniformity and uncertainty of complex systems. Therefore, it can be used as an index to measure the complexity of the system. On the other hand, the fractal market hypothesis, which aims at the deficiency of the efficient market hypothesis, has played an important role in the efficiency analysis of financial markets in recent years. In this paper, entropy and efficiency index are taken as the measurement of market complexity and efficiency respectively, and the closing price data of Shanghai Composite Index from July 15th 1991 to June 29th 2012 are taken as the research object. According to the characteristics of the Shanghai Composite Index, it is divided into 7 time intervals, and the above problems are studied in the following aspects:. First of all, based on the absolute deviation of Hurst index and 0.5 to represent the efficiency of the market, and using multifractal to eliminate trend volatility analysis (MF-DFAA) empirical study of the effectiveness of the Shanghai stock market, the results show that, Shanghai stock market is far from reaching the "weak" effective state. Secondly, the complexity of Shanghai stock market is discussed by using permutation entropy theory. The empirical analysis shows that permutation entropy can extract the characteristics of all kinds of stock price change period in time, and reflect the order of stock price change. In order to correctly reflect the complexity of the stock market. Finally, the relationship between efficiency and permutation entropy of Shanghai stock market is analyzed by using "sliding window" technique. The BDS test shows that there is a nonlinear relationship between efficiency index and permutation entropy. The cross correlation coefficient analysis shows that there is a cross effect between the two, and the long-term effect is more significant than that in the short term, and the nonlinear Granger test method proposed by Diks et al. shows that there is a very robust bidirectional Granger causality between the two. Therefore, the increase of short-term market complexity may reduce market efficiency, but in the long run, we can improve the efficiency of stock market by increasing market complexity. That is, market efficiency and the complexity of the market itself is closely related.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】