Practical Application of the SIML Estimation of Covariance, Correlation, and Hedging Ratio with High-Frequency Financial Data
Practical Application of the SIML Estimation of Covariance, Correlation, and Hedging Ratio with High-Frequency Financial Data
复制标题
高频金融数据的协方差、相关性和对冲比率的 SIML 估计的实际应用
DOI：
10.1007/978-981-13-8311-3_5
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Hiroumi Misaki
中科院分区：
文献类型：
--
作者：
Hiroshi Gunji;Kazuki Hiraga;Kenji Miyazaki;羽方康恵;羽方康恵;Hiroumi Misaki;Hiroumi Misaki;Hiroumi Misaki
Kunitomo和Sato提出了分离信息最大似然（separating information maximum likelihood estimation of已实现波动率和协方差with micro-market noise， 2008 [12]; Math Comput SIML 8:1272-1289, 2011 [13]; N Am economics finance 26:282-309, 2013[14]）方法，用于利用市场微观结构噪声估计高频数据的综合波动率和协方差。通过渐近分析和大量蒙特卡罗模拟证明，SIML估计器即使在不规则、非同步和有噪声的数据下也具有合理的渐近性质和有限样本性质(Kunitomo et al. in Asia-Pac financial Markets 22(3):33 - 368, 2015 [11]；《经济研究》（英文版），2015年第1期。Misaki(通过高频交易和报价的SIML估计对波动性进行实证分析。施普林格，Cham, pp. 66-75[18])利用实际交易个股的数据，采用SIML估计对波动性进行了实证研究。在本研究中，我们分析了多变量高频金融数据，以检验SIML方法在估计综合协方差、相关性和套期保值比率方面的有效性。此外，我们通过比较基于估计对冲比率构建的简单投资组合的表现来测试对冲的效率。我们的研究结果表明，SIML估计对于分析来自实际市场的多变量高频数据以及单变量案例都是有用的。
The separating information maximum likelihood (SIML) method was proposed by Kunitomo and Sato (Separating information maximum likelihood estimation of realized volatility and covariance with micro-market noise, 2008 [12]; Math Comput Simul 8:1272–1289, 2011 [13]; N Am J Econ Financ 26:282–309, 2013 [14]) for estimating integrated volatility and covariance using high-frequency data with market microstructure noise. The SIML estimator has reasonable asymptotic properties and finite sample properties even with irregular, non-synchronized, and noisy data, as demonstrated by means of asymptotic analysis and massive Monte Carlo simulations (Kunitomo et al. in Asia-Pac Financ Markets 22(3):333–368, 2015 [11]; Misaki and Kunitomo in Int Rev Econ Financ 40:265–281, 2015 [19]). Misaki (An empirical analysis of volatility by the SIML estimation with high-frequency trades and quotes. Springer, Cham, pp. 66–75 [18]) conducted an empirical study on volatility by employing SIML estimation with data of actually traded individual stocks. In the present study, we analyze multivariate high-frequency financial data to examine usefulness of the SIML method for estimating integrated covariance, correlation, and hedging ratio. Additionally, we test the efficiency of hedging by comparing the performances of simple portfolios constructed based on estimated hedging ratios. Our findings suggest that SIML estimation is useful for analyzing multivariate high-frequency data from actual markets as well as univariate cases.