作　　者： ; ; ;

机构地区： 广东金融学院金融系

出　　处： 《广东财经大学学报》 2016年第4期81-88,共8页

摘　　要： 为高频交易数据构造的价差指标寻找低频的替代性指标是当前金融学研究的热点问题之一。现有研究表明,Corwin和Schultz构造的高低价差指标在刻画买卖价差的横截面特性方面要优于其他指标,但该指标能否用作资产定价研究中买卖价差的代理变量还没有得到相应的检验。以A股上市公司1992年-2014年间的数据为样本,检验高低价差指标和期望收益率之间的关系,研究发现：无论是在常数收益率模型的框架下还是在CAPM模型的框架下,抑或是在Fama和French模型的框架下,高低价差指标都与期望收益率不相关,且股权分置和＂季节效应＂都不是两者之间不相关的原因。这在一定程度上说明高低价差指标不能作为资产定价中买卖价差的代理变量。 Finding a low-frequency variable as a proxy indicator for the bid-ask spread constructed for the high-frequency transaction data is one of the hot topics in current financial researches. Current researches indicate that the high-low spread constructed by Corwin and Schuhz （2012） is better in describing cross-sectional bid-ask spread, but whether the high-low spread can be used as a proxy variable in asset pricing researches has not yet been proved. Under the framework of Farna and French （ 1992 ） , this paper examines the relationship between the high-low spread and the expected rate of return using the sample of all A-share companies between 1992 and 2014. The research results show that, no matter whether it is under the CAPM model framework or under the Fama and French （1992） framework, the relationship between the high-low spread and the expected rate of return is insignificant, and the non-tradable share reform and the seasonal effect are not responsible for such an insignificance. To some degree, these results indicate that high-low spread cannot be used as a proxy variable for bid-ask spread in asset pricing.

关 键 词： 模型 买卖价差 高低价差 期望收益率 代理变量 资产定价 高频交易 股票流动性

领　　域： [经济管理—金融学]