重大事件下中国股市跳跃行为特征分析

发布时间：2018-06-16 13:54

本文选题：重大事件 + 跳跃性　；参考：《南京财经大学》2012年硕士论文

【摘要】：股票市场收益率通常是小幅波动的，但是当市场出现重大事件或者异常信息时，在短时间内收益率极有可能会发生大规模的运动，产生跳跃性变化，市场波动率也明显加剧。这对投资者、风险管理部门、政策决策部门都是极为不利的，事态严重的甚至可能导致经济遭受巨大损失，公司破产，市场萎靡等许多危害。因此研究重大事件对中国股票市场的影响，，对提高金融机构和风险管理部门的风险管理水平，进行及时有效的风险监控和防范，对金融监管部门相关政策制定等都具有一定的现实指导意义。既然重大事件突发会引发市场剧烈波动，从而导致经济体受到严重创伤，基于此本文针对重大事件对市场影响的研究提出了几个问题：中国的股票市场中是否存在跳跃现象？存在的跳跃变化点与重大事件的发生存在怎样的对应关系？当重大事件发生的时候，中国股票市场会表现怎样的跳跃特性和波动特性？基于上述问题，本文检验了中国股票市场收益率的跳跃性、波动时变性以及长期记忆性三大特性的存在性。同时考虑到了这三大特性的市场拟合模型将更加贴近现实，实用价值将更高。这也克服了现有相关文献研究的两大不足：其一，在构建符合中国股票市场的模型时仅仅考虑了三大特性中的跳跃性或者波动性，弥补了对二者的复合特性的研究空缺。其二，针对中国股市市场受重大事件影响的研究，至今仍缺乏对重大事件较为系统的研究。本文就此研究了中国股票市场近十几年来跳跃变化点与重大事件的对应关系，表明了研究重大事件对市场影响的必要性，同时对近十三年来发生的重大事件进行分类，主要有政治、政策、经济及自然灾害四类，然后从跳跃后验概率、跳跃强度、跳跃大小均值等角度研究各类性质重大事件的跳跃特性，分析了重大事件对市场影响的滞后期、跳跃频率以及跳跃幅度，为了更加全面的反映重大事件对市场波动率的影响，文章引入波动成分探讨了各类性质的重大事件对市场波动率的长期与短期持续时间与影响程度。研究表明不同性质的重大事件影响各异，即使同一性质的重大事件影响程度也差异甚大。这些研究对投资者或投资机构、金融风险管理部门、政策制定者、市场监管部而言，可以根据历史上重大事件的影响程度，来设立相应的机制实现事前预警、事时度量和事后管理，从而把市场风险控制在有效承受的范围内。
[Abstract]:The stock market returns are usually small fluctuations, but when there are important events or abnormal information, in a short period of time, it is very likely that large scale movement will occur, resulting in jump changes, market volatility is also significantly increased. This is extremely disadvantageous to investors, risk management departments, and policy decision making departments. Serious events may even lead to huge losses to the economy, bankruptcy of companies, weak markets, and many other hazards. Therefore, to study the impact of major events on China's stock market, to improve the risk management level of financial institutions and risk management departments, and to carry out timely and effective risk monitoring and prevention, It has certain practical guiding significance to the relevant policy formulation of the financial supervision department. Since the sudden occurrence of major events can lead to severe market fluctuations, which leads to severe trauma to the economy, this paper puts forward several questions on the study of the impact of major events on the market: is there a jump phenomenon in the stock market of China? What is the relationship between the jumping point of existence and the occurrence of major events? When the major events occur, how will the Chinese stock market performance of the jump and volatility characteristics? Based on the above problems, this paper examines the existence of three characteristics of the Chinese stock market: jump, volatility, and long-term memory. At the same time, considering the three characteristics of the market fitting model will be closer to the reality, practical value will be higher. This also overcomes the two shortcomings of the existing literature. Firstly, in constructing the model which accords with the Chinese stock market, we only consider the jump or volatility of the three characteristics, and make up for the lack of research on the compound characteristics of the two. Second, there is still a lack of systematic research on major events in China's stock market. In this paper, we study the corresponding relationship between the jumping point and the major events in the Chinese stock market in the past ten years, and show the necessity of studying the influence of the major events on the market. At the same time, we classify the major events that have taken place in the past 13 years. There are four main categories: politics, policy, economy and natural disasters. Then the jump characteristics of major events of various nature are studied from the angles of jump posteriori probability, jump intensity, and mean jump size, and the lag period of the influence of major events on the market is analyzed. In order to reflect the influence of major events on market volatility more comprehensively, this paper introduces volatility components to explore the long-term and short-term duration and degree of influence of major events on market volatility. The results show that the influence of major events of different properties is different, even if the influence degree of major events of the same nature is very different. For investors or investment institutions, financial risk management departments, policy makers, and market regulatory departments, these studies can set up appropriate mechanisms for pre-warning based on the extent to which historical events have affected them. Time measurement and ex post management to control market risk within the range of effective exposure.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【参考文献】