中国股市河北板块的分形特征研究

发布时间：2018-05-22 16:47

本文选题：河北板块 + R/S分析　；参考：《河北经贸大学》2012年硕士论文

【摘要】：股票作为一种常用的金融交易工具,其价格波动问题是当前金融研究中的一项基本且重要的课题。有效市场理论在描述股票市场时被发现带有一定的局限性。分形理论借助于非线性系统来分析股票市场复杂的波动状态,解决了有效市场理论中许多前提假设的局限性和缺陷,从而所解释股票市场的结构和价格波动特性更加接近于真实市场特性。因此,本文运用分形理论对股票市场结构和股价波动特征进行研究。首先,通过对所研究的数据进行正态性检验、相关性检验和非线性检验,验证了河北板块股票收益率数据是非正态的、非线性的,从而证实了运用有效市场假说分析股票市场确实存在局限性。同时,这也为分形方法的应用提供了条件。其次,对河北板块股票收益率进行单分形分析,发现股票收益率存在循环周期(即具有长期记忆性),并从整个收益率序列、循环周期前和循环周期后三个时间段来计算Hurst指数。结果表明,H值具有明显的差别：记忆周期前的H值整个时间区间的H值记忆周期后的H值,从而证实河北板块股票收益率存在分形特征。另外,上证股票走势比深证股票的持续性趋势更明显,相对风险较小。再次,在存在单分形特征的基础上,通过广义Hurst指数、多重分形谱来检验收益率序列是否存在多重分形特征。结果表明,收益率序列确实存在多重分形特征,并且分形特征是由收益率序列的自相关性和胖尾分布引起的。同时也说明,单分形理论并不能对股票收益率序列进行很好的描述,在结论的分析上应以多重分形结果为依据。最后,针对河北板块股票收益率存在分形特征的现象,从河北板块和全国股市大环境去探寻其原因。认为,投资者中个体投资者占比重大且对信息的反应不一致,是产生分形特征的主要原因；上市公司信息披露的不真实、不及时、不充分；国有企业改制上市后认识的偏差、体制上的欠缺和操作上的不当；股票市场本身的开放性和非线性；上市企业发展和成长历程的不同以及发展和成长历程的非线性也都是导致股票收益率出现分形特征的原因。据此,针对证券监管者、上市公司、投资者三方面给出了合理化建议。
[Abstract]:As a common financial trading tool, stock price fluctuation is a basic and important subject in current financial research. The efficient market theory is found to have some limitations in describing the stock market. Fractal theory analyzes the complex fluctuation state of stock market by means of nonlinear system, and solves the limitation and defect of many premise hypotheses in efficient market theory. Thus, the structure and price fluctuation characteristics of the stock market are closer to the real market characteristics. Therefore, this paper uses fractal theory to study the structure and volatility of stock market. Firstly, through normality test, correlation test and nonlinear test, the paper verifies that the data of stock yield of Hebei plate are non-normal and nonlinear. It is proved that the analysis of stock market by efficient market hypothesis has its limitations. At the same time, this also provides the condition for the application of fractal method. Secondly, the single-fractal analysis of the stock yield in Hebei plate shows that there exists a cycle (that is, long-term memory) in the stock return rate, and the Hurst exponent is calculated from the whole yield sequence, the period before and after the cycle. The results show that the H value has obvious difference: the H value before the memory period is the H value of the whole time interval after the H value memory period, thus confirming the existence of fractal characteristics of the stock yield in Hebei plate. In addition, the Shanghai stock market trend than Shenzhen stock market trend is more obvious, relatively less risk. Thirdly, on the basis of the existence of single fractal features, the multifractal features of yield series are tested by generalized Hurst exponent and multifractal spectrum. The results show that the multifractal feature does exist in the yield series, and the fractal feature is caused by the autocorrelation and the fat tail distribution of the return sequence. It also shows that the single-fractal theory can not describe the stock return series well, and the conclusion should be based on the multifractal results. Finally, in view of the phenomenon that the stock yield of Hebei plate has fractal characteristics, this paper explores the reasons from the stock market environment of Hebei plate and the whole country. It is believed that individual investors account for a large proportion of investors and react inconsistently to information, which is the main reason for fractal characteristics; the information disclosure of listed companies is not true, timely and insufficient; The lack of system and improper operation; the openness and nonlinearity of stock market itself; the difference of development and growth history of listed enterprises and the nonlinearity of development and growth process are also the reasons for the appearance of fractal characteristics of stock yield. Accordingly, for securities regulators, listed companies, investors three aspects to give reasonable advice.
【学位授予单位】：河北经贸大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F832.51

【参考文献】