基于价格极值构建有效价差的广义矩估计

发布时间：2018-03-15 23:06

本文选题：流动性　切入点：买卖价差　出处：《运筹与管理》2017年03期 　论文类型：期刊论文

【摘要】：基于Corwin和Schultz(2012)提出的有效价差的High-Low估计,结合价格极值信息得到新的一阶矩条件,构造了有效价差的广义矩估计。随后通过随机数值模拟比较了基于价格极值的广义矩估计(GMM)与Roll的协方差估计、Bayes估计以及Corwin和Schultz的High-Low估计在多种不同状态下的估计精度。数值模拟结果显示,无论在交易连续的理想状态下还是交易不连续且波动率相对不高的非理想状态下,GMM估计的精度均高于其余三种估计;基于我国股票市场的实例分析,也表明GMM估计的估计精度优于其余三种估计。因此,GMM估计为度量金融资产的交易成本提供了一种有效方法。
[Abstract]:Based on the High-Low estimation of the effective price difference proposed by Corwin and Schultzian 2012, a new first-order moment condition is obtained by combining the price extremum information. The generalized moment estimation of effective price difference is constructed, and then the estimation of generalized moment estimator based on price extremum is compared with that of Roll's covariance estimate and High-Low estimate of Corwin and Schultz in different states by random numerical simulation. Accuracy. Numerical simulation results show that, The accuracy of GMM estimation is higher than that of the other three estimators, whether in the ideal state of continuous trading or in the non-ideal state of trading discontinuity and relatively low volatility. It also shows that the accuracy of GMM estimation is better than that of the other three estimators, so it provides an effective method to measure the transaction cost of financial assets.
【作者单位】：北京工业大学经济与管理学院;
【基金】：国家自然科学基金资助项目(61273230,61603010,61603011) 中国博士后科学基金项目(2015M580033,2015M580940)
【分类号】：F831.51

【参考文献】