论文部分内容阅读
本文基于半鞅过程和非参数统计推断方法,利用已实现幂变差的渐进统计特性,构造检验统计量,在统一的分析框架下,对金融资产价格中随机波动、跳跃和微观结构噪声等问题进行全面系统的研究。并根据上海证券交易所不同行业的股票,上证50股票指数及其成分股的高频数据进行实证研究。结果表明,我国A股市场中,噪音交易显著;约43%的风险来源于资产收益过程的随机波动风险,可用股票期权交易对冲;不同来源风险的重要性程度依次为:随机波动的风险、系统性跳跃风险以及异质性跳跃风险;流动性越好的股票越显示出跳跃、尤其是无限小跳的证据。
Based on the semi-martingale process and the non-parametric statistical inference method, this paper uses the asymptotic statistical properties of power variation to construct the test statistic. Under the unified analysis framework, this paper analyzes the problems of stochastic volatility, jump and microstructure noise in the financial asset price Conduct a comprehensive and systematic study. And based on the Shanghai Stock Exchange stocks in different industries, the Shanghai Stock Exchange 50 stock index and constituent stocks of high-frequency data for empirical research. The results show that noise trading is significant in China’s A-share market. About 43% of the risk comes from random fluctuations of asset returns and can be hedged with stock options. The importance of different sources of risk is in turn: risk of stochastic volatility, Risk of sexual jumping, and risk of heterogeneous jumping; stocks with better liquidity show more jumps, especially with infinitesimal jumps.