动量效应实证研究在中国股市的特殊性

发布时间：2018-07-10 07:47

本文选题：动量效应 + 现金分红　；参考：《复旦大学》2012年硕士论文

【摘要】：本文目的在于通过验证和分析沪深300指数样本股票中动量效应的存在,探索中国A股市场分红政策、个股市场权重分布的特殊性对于动量策略效果的影响。首先,本文介绍了相关传统金融经济学理论,包括有效市场假设(EMH)、传统CAPM模型及其扩展。参考Jegadeesh和Titrman(1993)中的方法,我们构建投资策略J/K/Type,其中J和K分别代表"Sorting Period"和"Holding Period",而参数Type的设定是为了研究现金分红对于策略J/K结果的影响,可以取0和1,Type取值的不同主要是为了区分策略构建中使用个股收益率的差异。我们使用CSMAR中国股票市场数据库中2005年5月到2011年12月的个股月收益率和收盘价,通过改变参数J和K的值,验证了动量效应在中国股市的存在性。接下来,本文结合沪深300指数编制细则、中国A股市场现状,分别从现金分红、个股权重和个股β系数的角度对不同参数的J/K/Type策略进行了分析。通过分析不同投资组合的月平均收益,我们并没有发现沪深300指数成份股中存在明显的动量效应,相反的,对于策略J/K,我们一致得到了"Losers beat Winners"的结论。另外,"Winners"组合和"Losers"组合的月平均收益随着J和K取值的改变也呈现一定的规律性："Sorting Period" J越短,"Holding Period" K越长,投资组合收益率高的可能性较大。本文还考虑到中国股票市场上存在的分红制度不规范、权重股严重等不同于欧美等发达国家成熟市场的特点,对中国股市不存在动量效应的论断提出质疑,提出动量效应的验证还需要考虑分红、个股权重、样本范围等其他因素的影响,并分别从实证和理论角度分析了个股权重与个股β系数之间的关系以及权重股对于其他股票β系数的影响。我们发现随着个股平均市场权重的增大,个股平均β系数呈现先增大后减小,而且当个股权重达到一定比例后,β系数小于1的概率随着成分股流通股在市场组合中权重的增加而增大。
[Abstract]:The purpose of this paper is to verify and analyze the existence of momentum effect in the sample stock ticket of CSI 300 index, and to explore the effect of the dividend policy and the special weight distribution of the individual stock market on the momentum strategy effect in China's A-share market. Firstly, this paper introduces the traditional financial economics theory, including efficient Market hypothesis (EMH), traditional CAPM model and its extension. Referring to the method in Jegadeesh and Titrman (1993), we construct the investment strategy J / K / Type. where J and K represent "sorting period" and "holding period" respectively, and the parameter Type is set to study the effect of cash dividend on the results of the strategy J / P K. The difference between 0 and 1T can be used to distinguish the differences in the return rate of individual stocks used in strategy construction. We use CSMAR Chinese stock market database from May 2005 to December 2011 to verify the existence of momentum effect in Chinese stock market by changing the values of parameters J and K. Then, combining with the detailed rules of Shanghai and Shenzhen 300 index and the current situation of China's A-share market, this paper analyzes the different parameters of J / K / Type strategy from the perspectives of cash dividend, weight of individual stock and 尾 coefficient of individual stock. By analyzing the average monthly returns of different portfolios, we do not find that there is a significant momentum effect in the CSI 300 index. On the contrary, for the strategy J / K, we all get the conclusion of "Los beat wins". In addition, the monthly average returns of "winners" and "Losers" also show certain regularity with the change of J and K values: the shorter the "sorting period" J, the longer the "holding period" K, and the higher the return probability of the portfolio. This paper also takes into account the characteristics of the non-standard dividend system and heavy stock market in China, which are different from the mature markets in developed countries such as Europe and the United States, and questions the conclusion that there is no momentum effect in China's stock market. It is pointed out that the validation of momentum effect also needs to consider other factors, such as dividend, weight of individual stock, sample scope, etc. The relationship between the weight of individual stock and the 尾 coefficient of individual stock and the influence of weight on the 尾 coefficient of other stocks are analyzed respectively from the perspective of empirical and theoretical analysis. We find that with the increase of the average market weight of individual stock, the average 尾 coefficient of individual stock increases first and then decreases. When the weight of each stock reaches a certain proportion, the probability of 尾 coefficient less than 1 increases with the increase of the weight of the tradable stock in the market portfolio.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【参考文献】