中国股票市场数字文化下的价格聚类研究

发布时间：2018-01-20 04:28

本文关键词： 价格聚类传统文化交易规则牛熊市　出处：《西南交通大学》2013年硕士论文　论文类型：学位论文

【摘要】：某些数字在价格中出现频率明显高于其它一些数字出现频率的现象被称为价格聚类现象,它广泛存在于各种金融市场中,既有理论多认为其发生的直接原因在于人为偏差和投资者的有限理性,并形成吸引假说与价格谈判假说等较完善的相关理论假说。吸引假说一般认为数字发生聚类可能性从大到小依次是0,5,{2=8},{3=7,4=6},{1=9}；价格谈判假说认为,聚类程度一般与股票的价格水平、波动性水平正相关,而与股票的流动性水平和公司规模负相关。价格聚类是否广泛且长期存在于中国股票市场?若存在,是否具有不同于其它市场的特征,是否能用既有假说加以解释?本文以中国股市为研究对象,融入文化因素,并结合市场发展实际,详细探讨中国股票市场上的价格聚类现象。研究发现,中国股市中价格聚类确实广泛且长期存在但并不完全服从吸引假说：与许多国外市场表现一致的是价格尾数一般在0处聚类最多,其次是5；但是最与众不同之处在于“喜8恶4”且不同市场、板块、股票种类和交易价格的聚类表现不尽一致。本文认为尾数8高频率而4低频率现象就是中国传统文化中对8喜爱而对4忌讳的充分显现。以价格谈判假说所涉影响因素为基础,本文引入交易规则和牛熊市两个市场整体影响因素以及中国五大传统节日(春节、清明节、端午节、中元节和中秋节)也作为影响因素,分市场、板块、股票种类全面分析“喜8恶4”聚类现象。研究发现深交所的“8vs4”的聚类现象受上述因素的影响较上交所更明显(特别是传统节日的影响),对A股的影响要较B股明显,传统节日和牛熊市对主板和中小板市场的影响也存在显著差异：A股较符合价格谈判假说但B股却不能用该假说很好地解释；2006年7月1日新交易规则的实施使得聚类发生概率普遍降低；牛市时的聚类发生概率普遍高于熊市；传统节日因素对该种聚类的显著影响确实存在但并不普遍,相对来说在A股市场上的影响要比在B股市场上更普遍。本文最后给出一个利用前文所研究的聚类现象可能进行未来收益率预测的简单模型,作为实际应用的一个简单尝试。这使得本文的研究不仅丰富了价格聚类的研究,同时对投资决策制定等方面也具有积极的意义。
[Abstract]:The phenomenon that some numbers appear more frequently in the price than others is called price clustering phenomenon, which widely exists in all kinds of financial markets. Most of the existing theories think that the direct cause of its occurrence lies in the artificial deviation and the limited rationality of investors. The attractive hypothesis, the price negotiation hypothesis and the other relatively perfect related theoretical hypotheses are formed. It is generally believed that the probability of numerical clustering is 0 / 5, {2 / 8} and {3 / 7 / 4 / 6}, respectively. {1 / 9}; The price negotiation hypothesis holds that the clustering degree is generally positively correlated with the stock price level and volatility level, but negatively correlated with the stock liquidity level and firm size. Does price clustering exist widely and for a long time in Chinese stock market? If so, are they different from other markets and can they be explained by existing hypotheses? This paper takes the Chinese stock market as the research object, integrates the cultural factors, and discusses the price clustering phenomenon in the Chinese stock market in detail in combination with the market development practice. It is found that the price clustering in the Chinese stock market is indeed widespread and long-existing, but it is not completely satisfied with the attraction hypothesis: in accordance with the performance of many foreign markets, the price Mantissa is generally the most clustered at zero, followed by 5; But the most distinctive thing is that the "likes 8 and evil 4" and different markets, plates. This paper holds that the phenomenon of high frequency of Mantissa 8 and low frequency of 4 is the full manifestation of love of 8 and taboo of 4 in Chinese traditional culture. Based on the influencing factors of the price negotiation hypothesis, this paper introduces the trading rules and the overall influence factors of bull market and bear market, as well as the five traditional festivals of China (Spring Festival, Ching Ming Festival, Dragon Boat Festival). Ghost Festival and Mid-Autumn Festival) also as the influence factor, sub-market, plate. The study shows that the cluster phenomenon of "8vs4" in Shenzhen Stock Exchange is more affected by the above factors than by the Shanghai Stock Exchange (especially the influence of traditional festivals). The impact on A shares is more obvious than that on B shares, and there are significant differences between traditional festivals and bull bear markets on the main and small board markets: the A shares are more in line with the price negotiation hypothesis, but the B shares cannot be well explained by this hypothesis; In July 1st 2006, the implementation of new transaction rules reduced the probability of clustering. The probability of clustering in bull market is higher than that in bear market. The significant influence of traditional festival factors on the clustering is real but not universal, and it is more common in A-share market than in B-share market. At the end of this paper, we give a simple model to predict the future rate of return by using the clustering phenomenon studied in the previous paper. As a simple attempt of practical application, the research in this paper not only enriches the research of price clustering, but also has positive significance for investment decision making and so on.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F832.51

【参考文献】