基于计算实验的中国股票市场停牌有效性研究

发布时间：2018-04-27 10:25

本文选题：金融市场 + 计算实验金融学　；参考：《天津大学》2014年硕士论文

【摘要】：停牌是为了加强信息披露、抑制股票的异常波动、缓解订单极度不平衡的情况保证证券交易有序进行而设立的一个重要的市场稳定机制，并广泛应用于全球主要证券市场。但停牌是否真的能稳定市场，其作用是具争议的。为了研究中国股票市场停牌机制的实施效果，本文采用极端事件统计分析方法进行实证研究以及应用计算实验金融的创新性的研究方法对停牌机制进行仿真模拟检验其有效性，对停牌有效性研究领域进行补充和创新。首先实证部分本文使用沪深300指数成分股中的深市股票从2009年8月到2011年8月的停复牌数据和高频交易数据，创新性地运用统计物理学中极端事件的统计分析技术动态展现停牌事件引起的市场价格变化和市场波动率演化过程，并将停牌分为正负事件分别研究，实证分析了中国股票市场停牌制度的实施效果，检验其有效性。实证结果显示正负停牌事件的累积收益率在停牌前一直持续增长或下降，而停牌后都趋于一个稳定值，停牌抑制了价格的持续增长或下降，起到了反转和回归的效果，表明停牌具有一定的价格发现作用。而停牌实施前后，绝对收益率出现了尖峰和尖峰后衰退的变化趋势，在复牌时刻达到波动最大值，之后逐渐衰退至非停牌日同期水平，衰退过程具有显著的幂律衰减性。结果表明停牌并未降低价格波动反而导致增加，停牌制度的实施并没有实现稳定市场的既定目标。为了研究停牌导致价格波动出现如此变化模式的原因，本文接着从订单簿不平衡角度出发，将清除订单簿作为停牌机制，利用计算实验金融的方法模拟市场运行，研究是否是订单簿不平衡导致的市场变化。在mason平台上加入停牌机制作为停牌仿真平台，并采用实证研究中的极端事件统计分析方法对仿真数据进行分析。结果显示仿真结果与实证结果一致，停牌的确导致市场波动变大，，价格波动呈现尖峰衰退模式，衰退过程也具有幂律衰减性。计算实验较好地模拟出了停牌机制下的市场的真实动态，为此本文用计算实验解释实证中停牌导致市场发生如此变化的原因，认为是订单极度不平衡导致的停牌后价格变化出现尖峰以及幂率衰减的模式。
[Abstract]:The suspension of trading is an important market stability mechanism which is established to strengthen information disclosure, restrain the abnormal fluctuation of stock, alleviate the extreme imbalance of orders and ensure the orderly conduct of securities trading, and is widely used in the major securities markets in the world. But whether suspension really stabilizes the market, its role is controversial. In order to study the effect of the suspension mechanism in Chinese stock market, this paper uses the extreme event statistical analysis method to carry on the empirical research, and uses the innovative research method of computational experimental finance to carry on the simulation simulation to the suspension mechanism to verify its validity. To the suspension validity research area carries on the supplement and the innovation. First of all, the empirical part of this paper uses the Shenzhen Stock Exchange Stock Exchange data from August 2009 to August 2011 and high-frequency trading data. Innovatively using the statistical analysis technology of extreme events in statistical physics to dynamically show the market price change and market volatility evolution process caused by the suspension event, and divide the suspension into positive and negative events, respectively. This paper empirically analyzes the effect of the suspension system in China's stock market and tests its effectiveness. The empirical results show that the cumulative yield of positive and negative suspensions continues to increase or decrease before the suspension, but after the suspension tends to a stable value, the suspension inhibits the continuous growth or decline of prices, and has the effect of reverse and regression. It shows that the suspension has a certain price discovery function. Before and after the suspension of trading, the absolute rate of return appeared the trend of peak and post-peak recession, reached the maximum fluctuation at the time of resumption, then gradually declined to the level of the same period of the non-suspension day, the decline process has significant power law attenuation. The results show that the suspension does not reduce the price fluctuation but leads to an increase, and the implementation of the suspension system does not achieve the established goal of stabilizing the market. In order to study the reason of the price fluctuation caused by the suspension, this paper starts from the imbalance of the order book, takes the order book as the suspension mechanism, and simulates the operation of the market by the method of calculating the experimental finance. Study whether the market changes caused by the imbalance in the order book. The suspension mechanism is added to the mason platform as the simulation platform, and the statistical analysis method of extreme events in the empirical research is used to analyze the simulation data. The results show that the simulation results are consistent with the empirical results. The suspension of trading does cause the market volatility to become larger, the price fluctuations show a peak recession mode, and the decline process has power law attenuation. The calculation experiment simulates the real market dynamics under the suspension mechanism. Therefore, this paper uses the calculation experiment to explain the reason why the market changes so in the case of suspensions. It is considered to be the mode of peak price change and power rate attenuation after the suspension caused by the extreme imbalance of orders.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【参考文献】