【作者】 姚宁；

【导师】 王春峰；

【作者基本信息】 天津大学 ， 金融工程， 2009， 博士

【摘要】 证券市场波动性的研究长期以来一直是现代金融领域研究的主要问题之一,是证券二级市场的核心功能——价格发现与资本配置的核心。因此,在存在跳跃和市场噪音的条件下如何准确衡量资产均衡价格的波动性具有重要的理论和应用价值。本文在金融市场微观结构理论视角下沿用“已实现”波动研究的主流框架,从理论和实证角度研究了考虑跳跃和市场微观结构噪音条件下资产均衡价格波动性的估计方法以及均衡价格波动性估计方法在资产配置中的应用。全文从五个方面进行了分析探讨:1、市场微观结构噪音条件下中国股市已实现波动率跳跃行为的特征揭示。首先,以二次幂变差的测量为理论基础,将日间已实现波动率分解为连续样本方差和离散跳跃变差,研究了离散跳跃变差序列的统计特征,随后使用HAR-RV-CJ模型分别考察了连续样本方差和离散跳跃变差对已实现波动率预报的影响。2、在我国股市日间波动存在跳跃特征的前提条件下,建立了基于跳跃扩散过程的股票和股票指数间的事件风险模型,计算了我国大盘股和沪深300指数间的同时跳跃强度,分析了我国大盘股和沪深300指数可能存在的跳跃溢出效应。由于沪深300指数是我国即将推出的股指期货的标的指数,研究结果表明我国股市和即将运行的股指期货市场存在极强的风险关联性,为我国股市和期市跨市场监管的机制设计奠定了基础。3、中国证券市场微观结构噪音的特征研究。使用完全信息交易成本(FITC)这一代理指标测量了交易价格与完全信息价格的偏离程度,考虑了有效价格和完全信息价格的差异。研究结果表明完全信息交易成本考虑了私人信息和市场参与者的学习过程,准确估计了实际的交易成本,是市场效率的理想测量方式。4、在市场微观结构噪音与跳跃同时存在条件下资产均衡价格波动率的估计研究。应用时间序列变点的小波分析方法,对包含市场微观结构噪音和跳跃的高频价格采样数据的跳跃变差和积分波动率进行了有效的估计。研究结果发现小波分析方法能够准确识别我国股市的价格跳跃,提高波动性的估计精度,表明在金融市场中应当使用均衡价格波动性进行风险管理和资产配置。5、市场微观结构噪音条件下资产均衡价格波动率估计的应用研究。从投资者资产配置的角度探讨了市场微观结构噪音条件下资产均衡价格波动率估计的重要性。研究结果发现使用最优采样频率方法计算的“已实现”方差和“已实现”协方差能够显著提高投资者的经济收益,表明均衡价格波动率估计具有非常广阔的应用前景。更多还原

【Abstract】 Research on securities market volatility has been the main issue in the field of modern financial research for a long time. The volatility is the core of the function of securities market, which is price discovery and capital allocation. Therefore, how to estimate volatility of asset equilibrium price considering jump and market microstructure noise has important theoretical and applied value. In this paper, we study the estimation method of volatility of asset equilibrium price considering jump and market microstructure noise and the application of asset allocation from the perspective of financial market microstructure along realized volatility research framework. The dissertation includes five theoretical and empirical aspects:1. Features research of jump behavior of realized volatility in Chinese stock market considering market microstructure noise. Initially, building on theoretical results for bi-power variation measures, realized volatility is divided into two parts: the continuous sample path variation and the discontinuous jump variation. The statistical feature of jump variation is studied. Then how continuous sample path variance and discrete jump variation affect the prediction of realized volatility respectively is examined by using a HAR-RV-CJ volatility forecasting model.2. Under the precondition of the existence of jump in Chinese stock market, the event risk model based on the jump diffusion process is established for stock and stock index. We calculate the simultaneous jump intensity for large-cap stocks and Hushen 300 index. Risk spillover effect is analyzed between large-cap stocks and Hushen 300 index. The Hushen 300 index is the subject index of stock index futures coming soon, so the result shows that there is strong risk relationship between our stock market and stock index futures market. Finally the foundation is laid for united monitoring mechanism designing between stock market and stock index futures market in China.3. Features research of market microstructure noise in China. We measure the difference of transaction price and full information price by using an acting indicator which is named full information transaction cost. The difference between efficient price and full information price is considered by full information transaction cost. The results show that full-information transaction cost includes private information and learning on the part of market participants. As an ideal measurement of market efficiency, transaction cost can be accurately estimated through full-information transaction cost.4. How to estimate the volatility of asset equilibrium price is studied under the condition of jump and market microstructure noise existing at the same time. Based on the wavelet method of time series change-point analysis, jump variation and integrated volatility is consistent estimated using high frequency price sampling data which contains market microstructure noise and jump. We find that price jump of Chinese stock market is accurately identified by wavelet transform method and volatility is estimated more accurately, indicating that the volatility of asset equilibrium price should be used for risk management and asset allocation in financial market.5. Application research of estimation method of asset equilibrium price volatility considering market microstructure noise. The importance for estimation of asset equilibrium price volatility considering market microstructure noise is analyzed from the perspective of asset allocation. We find that the gains yield by optimal sampling realized variance and covariance are economically large, indicating that estimation of asset equilibrium price volatility has broad prospects.更多还原