高频金融数据中市场微观结构噪音误差估计  

Estimation of Market Microstructure Noise Error under High-frequency Financial Data

作　　者：叶绪国[1] 杜雪樵[2] 

机构地区：[1]凯里学院数学科学学院,凯里556011 [2]合肥工业大学数学学院,合肥230009

出　　处：《大学数学》2012年第5期62-69,共8页College Mathematics

基　　金：贵州省教育厅自然科学基金(黔教科20090080;2010076);凯里学院校级课题(Z1214);凯里学院重点学科建设项目(KZD2009001)

摘　　要：由于在波动率估计中高频数据的使用,市场微观结构噪音的干扰对无偏的和一致的估计波动率已经变成了一种障碍.为了更好地估计真实波动率,噪音方差估计显得日益重要.本文基于目前关于波动率估计研究成果,提出了在不同的假设情况下估计市场微观结构噪音误差的方法,并与常用的估计方法进行深入的比较,得到它们的渐近性质,并且进行广泛的模拟研究它们的有限样本性质,并得到较有意义的结果.Due to the use of high-frequency data in the estimation of volatility, the interference of market microstructure noise has become an obstacle to the unbiased and consistent estimation of volatility. In order to better estimate the real volatility, the estimation of noise variance appears to he increasingly important. Based on the current research results on the estimation of volatility, we propose several kinds of the estimation method of market microstructure noise error under different assumptions and deeply test between them and old ones. We derive its asymptotic properties and perform extensive simulations to study the finite sample properties, and more meaningful results are obtained.

关 键 词：高频数据 市场微观结构噪声误差 估计 

分 类 号：O211.67[理学—概率论与数理统计]