我国股市的微观结构噪声与波动特征研究

发布时间：2018-04-03 14:28

本文选题：微观结构噪声　切入点：双时间频度之实现方差　出处：《暨南大学》2013年硕士论文

【摘要】：一般来讲，资产收益率的波动包括连续性波动和跳跃性波动。波动性既是投资者投资行为综合作用的结果，又是对当前市场运行状况的综合反映。但是，由于市场微观结构的作用，使得观测到的价格常常会由于交易过程的影响而偏离均衡价格，因此通过观测价格估计得到的波动性不能较为准确的反映市场运行的真实状况。交易过程中导致观测价格偏离均衡价格的各种因素即为微观结构噪声，包括非同步交易、市场参与者的流动性需要、交易者的执行策略、价格的离散变化、信息非对称等。因此，在考虑微观结构噪声的影响之后再去估计收益率的波动性能够提高估计的有效性，能够更加准确的反映证券市场的价格行为及证券市场的稳定性与效率状况，对于企业进行资产定价以及投资者进行资产配置都具有重要意义。本文是以沪深300指数每五分钟交易数据作为样本数据，，研究了次贷危机前期，当期与后期我国股市的微观结构噪声以及连续性波动和跳跃性波动的集聚效应，规模效应，杠杆效应还有综合效应等波动特征。先利用跳跃扩散模型离散小波变换方法检测跳跃行为，并估计出跳跃行为的发生位置，次数，幅度和方差；然后将跳跃性波动从观测数据中分离出来；再利用TSRV方法对调整后的数据建模，估计出我国股市的噪声和连续性波动，从而就分别得到了我国股市的连续性波动和跳跃性波动。最后根据估计结果，利用修正EGARCH模型对我国股市波动特征主要是集聚效应，规模效应，非对称效应以及综合效应进行研究。
[Abstract]:Generally speaking, the volatility of asset return includes continuous volatility and jump volatility.Volatility is not only the result of the comprehensive action of investor's investment behavior, but also the comprehensive reflection of the current market operation.However, due to the effect of market microstructure, observed prices often deviate from equilibrium prices due to the effects of the trading process.Therefore, the volatility obtained by price estimation can not accurately reflect the real situation of market operation.In the course of trading, the factors that cause the observed price to deviate from the equilibrium price are microstructure noise, including asynchronous trading, the liquidity needs of market participants, the execution strategy of traders, the discrete change of price, the asymmetric information and so on.Therefore, estimating the volatility of return after considering the influence of microstructural noise can improve the effectiveness of the estimation, and can more accurately reflect the price behavior of the securities market and the stability and efficiency of the securities market.It is of great significance for enterprises to price assets and investors to allocate assets.Based on the data of Shanghai and Shenzhen 300 index trading every five minutes, this paper studies the microstructural noise, the agglomeration effect and scale effect of continuous volatility and jump volatility in China's stock market in the early, current and late stages of the sub-prime crisis.The leverage effect also has the fluctuation characteristic such as comprehensive effect.The jump behavior is detected by the discrete wavelet transform method of the jump diffusion model, and the location, frequency, amplitude and variance of the jump behavior are estimated, and then the jump wave is separated from the observed data.Then the TSRV method is used to model the adjusted data, and the noise and continuity volatility of Chinese stock market are estimated, and the continuity volatility and jump fluctuation of Chinese stock market are obtained respectively.Finally, according to the estimation results, we use the modified EGARCH model to study the agglomeration effect, scale effect, asymmetric effect and comprehensive effect of stock market volatility in China.
【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F832.51

【参考文献】