投资者情绪与股指关系的实证研究

【作者】 刘亮；

【导师】 田素华；

【作者基本信息】 复旦大学 ， 金融（专业学位）， 2013， 硕士

【摘要】 行为金融学理论在投资者行为和心理方面的研究成功的解释过很多股票市场的异像,因此受到人们的广泛关注。投资者情绪是刻画投资者心理的一个重要指标,因此也成为行为金融学研究的一个很重要的方面。在行为金融学的框架里,投资者情绪是决定资产价格和市场运行的一个非常重要的因素。中国股市市场己发展了20年,但在这段时间里,股票市场暴涨暴跌,投资者犹如坐过山车一样,情绪也跟着大起大落,时而兴奋,时而悲伤。股票市场大幅波动似乎难以用经济、货币等因素来合理解释,而这些与投资者心理有非常大的关系,也就是本文要研究的话题——投资者情绪。开展投资者情绪研究的意义在于：第一,有助于理解投资者行为如何影响股市波动,更好的理解股票市场运行机制。第二,有助于完善和丰富现有的行为金融学理论。第三,有助于投资者发现股票市场的“错误定价”,帮助投资者获得超额收益。第四,有助于监管者把握大众投资者的心理特征,以及这些心理变化对证券市场造成的影响,及时为监管者提供调控政策建议,避免股市大起大落。如何度量投资者情绪是本文最关键的一个问题,国内外学者提出了度量投资者情绪的显性指标和隐性指标,并验证过了这些指标可以作为投资者情绪指数。不过这些指标都有他们明显的缺陷,本文试图通过因子分析方法,利用单个的情绪代理指标构建投资者情绪综合指数,并用它来检验对未来股票指数收益的预测能力。通过实证研究,本文发现,单个情绪代理变量对股票指数收益的预测效果不甚理想。本文利用构建出来的投资者情绪综合指数(CIIS)的月度数据对申万大盘指数、申万中盘指数和申万小盘指数未来1个月、3个月及6个月的收益率进行回归后发现,投资者情绪综合指数能够较好的预测股票指数的收益。更多还原

【Abstract】 Because of its partial succeess in explaining some of those anomalies,behavioral finance has got substantial attention from academics and practitioners.The investor sentiment,one proxy of the investor psychology,also becomes an important factor in the study of the investor behavior and psychology.Chinese stock market has developed for20years, but during that time, the stock market slump, investors like a roller coaster, the mood also follow change radically, and sometimes happy, sometimes sad. Stock market’s volatility seems to be difficult to use economic, monetary and other factors to explain, and investor psychology is very important, is also in this paper to study the topic-investor sentiment.The meaning of investor sentiment research reads as follow:1,it can give hint to understanding investors’behavior.2,it is useful to learn the interaction of market returns,variability and trade volume;3,to which extant it is an opportunity to earn extra returns can be declosed;4,it is helpful for the government to take policy and control the market.As an abstract concept,investor sentiment needs accurate measurement,which is the primary problem we have to solve.Foreign and domestic researchers have given us a sum of incestor sentiment indicators that have been proved to be sentiment proxies.This thesis is going to judge the forecasting ablility of several sentiment indexes by using the econometrics method of regression.Besides,factor analysis is uesd to construct a compositive index of investor sentiment (CIIS)by synthesizing theselected sentiment indexes mentioned before so that the stock return can be better forecast using CIIS.Through empirical research, this paper found that, market turnover ratio can forecast future stock return effectively while closed-end fund discount rate,first day return of IPO, volume and China’s macroeconomic leading index.The compositive index of investor sentiment (CIIS) has better forecasting ablility on stock return which has been spported by our empirical results.更多还原