论文部分内容阅读
本文从理论上分析比较了两类有效买卖价差(effective bid-ask spread)估计的统计性质,即Roll的协方差估计(Roll,1984)和最近由Corwin and Schultz(2012)提出的基于最高价和最低价得到的估计。与以往文献中采用估计价差与基准价差的相关系数来衡量和比较不同估计的优劣表现的做法有所不同,本文通过推导并对比两种估计的偏差、均方误差及其在大样本下的性质,从而在理论上证明了基于最高价和最低价的价差估计精度的确高于Roll(1984)的估计,并通过随机模拟对上述结论进行了验证。
This paper theoretically analyzes and compares the statistical properties of two types of effective bid-ask spreads, Roll’s covariance estimate (Roll, 1984) and the recently proposed high-price and high-price bidding scheme proposed by Corwin and Schultz (2012) Lowest estimate. Compared with the previous literature that uses the correlation coefficient of the estimated spread and the benchmark spread to measure and compare the performance of the different estimates, this paper derives and compares the deviations, mean square errors, It is proved theoretically that the accuracy of spread estimation based on the highest price and the lowest price is indeed higher than that of Roll (1984), and the above conclusion is verified by stochastic simulation.