金融波动的赋权“已实现”双幂次变差及其应用

金融波动是金融研究中的热点问题。金融高频数据比低频数据包含了更丰富的日内收益波动信息,因此对金融高频时间序列的研究成为金融领域中备受关注的焦点。"已实现"波动是利用高频数据计算金融波动率的全新方法,目前在金融高频数据的研究中应用十分广泛,但它具有误差较大和不稳健的缺点,因此各种改进方法应运而生,其中"已实现"双幂次变差克服了"已实现"波动的不稳健的缺点。本文提出赋权"已实现"双幂次变差的概念,不但继承了"已实现"双幂次变差的稳健性,而且满足无偏性和最小方差性,通过理论证明和实证研究都表明其能够更准确的度量金融波动率。

Weighted Realized Bipower Variation of Financial Volatility and Its Application

Volatility is a hottopic in financial research.People pay more and more attention to the high frequency data in finance because it contains more volatility information of intraday return than low frequency data does.Realized volatility is a completely new method to calculate volatility of high frequency data, which is applied widely in the study of high frequency data in finance.There are many methods to improve on the realized volatility for it has shortcomings of bigerror and non-robust ness.Among these,only the realized bipower variation overcomes the shortcoming of non-robust ness.The concept of weighted realized bipower variation which is put forward in this article,is not only robust but also unbiased and efficient. The theorem poof and the demonstration study also show the same conclusion:it can measure the volatility more precisely.