我国股市的波动不对称性特性及其成因探究

发布时间：2018-04-29 17:21

本文选题：波动不对称性 + GARCH模型　；参考：《上海外国语大学》2014年硕士论文

【摘要】：股票价格波动是标示风险的重要指标，对于投资者风险与收益的分析、监管者的有效监管、上市公司股东收益最大化目标的实现，都有着至关重要的作用。研究股票市场波动的规律性，分析引起股票市场波动的成因，为监管者监管、投资者投资、上市公司生产经营活动提供可循的依据，是股票市场理论研究和经验分析的重要内容。因此，对股票市场波动性进行研究，不但具有重要的理论价值，而且具有重要的现实意义。关于股市波动的研究主要有三个方向-波动的长记忆性，聚集性和波动不对称性，近年来关于波动不对称性的研究逐渐增多，其重要性也日益突出，而这也将是本文的研究主题。本文首先回顾了国内外对于波动不对称的研究，总结了这些学者们的主要研究方法以及结论，根据国外的学者对各国股市的波动研究成果，波动不对称在西方各国股市中普遍存在，且表现为“利空”消息带来的波动对“利好”消息更大。国内的研究也证明我国股市中存在着波动不对称现象，同时，一些学者发现表现形式随着历史发展有所变化，在股市早期表现为“利好”消息比“利空”消息带来的波动更大，随着股市发展逐渐成熟转变为“利空”消息比“利好”消息带来的波动大，与国外股市相一致。对于波动不对称的成因研究较少，主流的理论是Black提出的“杠杆假说”与Campell提出的“波动反馈说”。本文研究中国股市波动不对称的特性，，并探求波动不对称的成因。由于波动不对称的特性会随着股市的发展而变动，虽然前人对我国股市波动不对称是否存在做过研究，仍然有必要对新的样本再做实证分析确认。金融时间序列普遍存在尖峰厚尾的特性，而对其建立的计量经济学模型中的误差项也常有着异方差性，这在以往的研究中已被普遍证明。因此实证分析时要选取与这些特性想符合的模型，GARCH类模型是在做此类研究时广泛采用的模型，本文中也使用GARCH类模型，并针对本文采用的样本数据进行调整，拟定了具体的模型方程，本文的模型可以概括称为AR-TARCH-GED模型。在建立模型对我国股市收益率数据进行实证分析后，证明我国股市中存在波动不对称现象。进一步探讨波动不对称的成因，首先对“杠杆假说”以及“波动反馈说”进行介绍，指出“杠杆假说”的局限性，特别是在我国股市的特殊环境下，“杠杆假说”不具备很好的解释能力。“波动反馈说”比杠杆假说的局限小，但它也无法解释波动中出现的波动反转（亦即“利好”消息比“利空”消息带来的波动更大）现象。然后提出了引起波动不对称的两个假设原因，分别是股市成熟程度、投资者心理。先利用前文实证分析中采用的计量模型，横向与外国股市波动、纵向将沪指样本分割成不同时间段进行比较，表明在成熟程度不同的股市中波动不对称的特性显著不同，同时意外发现市况对波动不对称性存在影响；然后介绍行为金融的著名理论“前景理论”，解释投资者在股价变动方向不同的情况下的风险决策变化，投资者的这一心理会引起相应的波动不对称。分析结果认为这两种假设都有可能是波动不对称的成因，至于其中的因果传导过程具体是怎样的，以及这两种因素或者其他未被提及的因素中哪种才是影响波动不对称的决定性因素，本文限于研究者能力与时间的限制不作进一步探讨。
[Abstract]:Stock price fluctuation is an important index to mark risk. It plays an important role in the analysis of investor's risk and income, the effective supervision of the supervisor and the realization of the maximization goal of the shareholders of the listed company. It studies the regularity of the stock market volatility, analyzes the causes of the volatility of the stock market, and regulates the regulators and investors. Investment, which provides basis for the production and operation of the listed companies, is an important part of the theoretical and empirical analysis of the stock market. Therefore, it is of great theoretical and practical significance to study the volatility of the stock market. The study of the stock market wave is mainly in three directions - the long memory of volatility In recent years, the research on volatility asymmetry is increasing, and its importance is becoming more and more prominent, and this will also be the subject of this study.
This paper first reviews the research on volatility asymmetry at home and abroad, summarizes the main research methods and conclusions of these scholars. According to the research results of foreign scholars on the volatility of the stock market, volatility is common in the stock market of the western countries, and the volatility of "Lei" message is more favorable to "good" news. At the same time, some scholars have found that the form of performance has changed with the development of history. In the early stage of the stock market, the "good" news is more volatile than the "sharp" news. With the gradual maturity of the stock market, the "good" news is better than "good". The volatility of the news is large and consistent with the foreign stock market. There are few studies on the causes of volatility asymmetry. The mainstream theory is the "leverage hypothesis" proposed by Black and the "volatility feedback" proposed by Campell.
This paper studies the characteristics of volatility asymmetry in China's stock market and explores the causes of volatility asymmetry. Because the characteristics of volatility asymmetry will change with the development of the stock market, it is still necessary to make an empirical analysis of the new samples, although the previous study on the existence of volatility asymmetry in China's stock market is still necessary. There are often Heteroscedasticity in the error terms in the econometric model established by the peak, which has been universally proved in previous studies. Therefore, the empirical analysis should be used to select the models that are in conformity with these characteristics. The GARCH model is a widely used model in this kind of research, and the GARCH model is also used in this paper. In addition, the model equations can be adjusted according to the sample data used in this paper. The model of this paper can be generalized as AR-TARCH-GED model.
After the empirical analysis of the stock market yield data in China, this paper proves the existence of volatility asymmetry in China's stock market. The causes of volatility asymmetry are further discussed. First, the "lever hypothesis" and "volatility feedback" are introduced, and the limitations of the "leverage hypothesis" are pointed out, especially in the special environment of China's stock market. The "leverage hypothesis" does not have a good explanatory power. "Volatility feedback" is less limited than the leverage hypothesis, but it does not explain the phenomenon of volatility in volatility (that is, "good" messages are more volatile than "Leo" messages). Then, the two hypothesis of volatility asymmetry is proposed, respectively. The market maturity, investor psychology. First, using the measurement model used in the previous empirical analysis, the horizontal and foreign stock market fluctuations, the longitudinal Shanghai index samples are divided into different time periods to compare, indicating that the characteristics of volatility asymmetry in the stock market with different maturity are significantly different, and at the same time, it is found that the market conditions are unsymmetrical to volatility. And then we introduce the famous theory of behavioral finance, "prospect theory", to explain the risk decision changes of investors in the different direction of the change of the stock price. The investor's psychology will cause the corresponding fluctuation asymmetry. The result of the analysis is that the two hypotheses are likely to be the cause of the volatility asymmetry, as for the causal transmission. What is the specific course of the guide process, and which of the two factors or other unmentioned factors is the decisive factor affecting the asymmetry of volatility, this article is limited to the limitations of the researchers' ability and time.

【学位授予单位】：上海外国语大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51;F224

【参考文献】