基于价格极值构建有效价差的广义矩估计

摘要

基于Corwin和Schuhz(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.%This paper combines the information from daily high and low prices and constructs the new estimators for the effective bid-ask spread by the Generalized Method of Moments(GMM)based on the High-Low estimator proposed by Corwin and Schultz (2012).Then the numerical simulation studies are conducted to compare the estimation accuracy between the GMM estimators based on price ranges and other three common spread estimators in the literature,namely the Roll covariance estimator,Bayes estimator and High-Low estimator.Simulation results reveal that whether in the ideal case that the prices can be observed continuously or in the realistic unideal case and the volatility is relatively small,GMM estimators are more accurate than the other three estimators.Furthermore,an empirical study in Chinese stock markets also demonstrates that the estimated accuracy of the proposed GMM estimators is better than the other three estimators.Therefore,the proposed GMM estimators provide an effective method to measure the trading costs of financial assets.