基于高频金融数据的已实现赋权中位数方法及其应用

发布时间：2018-01-03 05:32

本文关键词：基于高频金融数据的已实现赋权中位数方法及其应用　出处：《重庆理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：金融市场信息连续影响着证券市场价格的运动过程,因此数据采集的离散程度会导致信息丢失的程度。但随着存储和计算技术的快速发展,使采集市场实时交易的饱含丰富信息的高频数据成为可能,所以对高频数据的度量与应用逐渐成为金融分析领域的重要趋势之一。金融资产收益波动率的度量是风险评估的核心,它直接决定着对风险管理的有效识别。常见的风险度量方法是在值风险法(VaR),且计算在值风险(VaR)的模型也在不断完善,国内外学者对已有的在值风险法(VaR)博采众长,先后提出各种风险度量模型。本文旨在：通过理论分析寻找较优的风险度量模型；比较我国深市和沪市两个证券市场间的异同，，其创新点如下：结合“已实现波动率(RV)、已实现二次幂变差(BV)、已实现三次幂变差(TV)、已实现中位数波动率(MedRV)”理论,提出了包含已实现中位数波动率(MedRV)和已实现赋权波动率(RRV)的已实现赋权中位数波动率(WMedRV),并证明了该波动率的的无偏性和最小方差性;通过比较不同波动率的描述统计量,证明了该波动率具有良好的稳定性。最后将不同的已实现波动率度量方法应用于我国两个证券市场的VaR值计算。另外，为了能够通过对比来观察已实现赋权中位数波动率(WMedRV)方法的有效性,本文设计了经典的VaR方法在我国两个证券市场中的应用。
[Abstract]:Financial market information continuously affects the movement of the stock market price, so the discrete degree of data acquisition will lead to the degree of information loss. However, with the rapid development of storage and computing technology. It is possible to collect high-frequency data with rich information for real-time trading in the market. Therefore, the measurement and application of high-frequency data has gradually become one of the important trends in the field of financial analysis. The measurement of volatility of financial assets returns is the core of risk assessment. It directly determines the effective identification of risk management. Domestic and foreign scholars have put forward a variety of risk measurement models. The purpose of this paper is to find a better risk measurement model through theoretical analysis; Comparing the similarities and differences between Shenzhen and Shanghai stock markets, the innovations are as follows: Based on the theory of "realized fluctuation rate (RV), second power variation difference (BV), third power variation difference (TVV) and median fluctuation rate (MedRV)". The realized weighted median volatility rate (WMedRV) including realized median volatility (MedRV) and realized weighted volatility (RRV) is proposed. The unbias and minimum variance of the volatility are proved. It is proved that the volatility has good stability by comparing the description statistics of different volatility. Finally, different realized volatility measurement methods are applied to the calculation of VaR value in two stock markets in China. In addition, in order to observe the effectiveness of the weighted median volatility (WMedRV) method by comparison, this paper designs the application of the classical VaR method in two securities markets in China.
【学位授予单位】：重庆理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F830.9;F224

【参考文献】