机构投资者持股与应计质量相关性的资产定价——来自中国A股市场的证据

发布时间：2018-03-06 17:13

本文选题：机构持股　切入点：应计质量　出处：《系统管理学报》2011年04期 　论文类型：期刊论文

【摘要】：通过运用Panel Data(固定效应下)计量方法,系统地分析了影响应计质量(AQ)的内在因素,并着重分析了机构持股与AQ的相关性,从总样本和分等级样本综合来看,AQ与机构持股比例成显著负相关,表明,从整体上来看,对于那些盈余波动较大,AQ较差的公司股票,机构投资者予以了识别与回避。进一步,通过构建AQ的替代变量AQfactor,将其纳入三因子和四因子模型,采用时间序列OLS回归计量方法进行资产定价检验。结果显示,AQfactor虽然与股票回报率正相关,但是,在三因子模型中,其回归系数只在10%水平上显著,同时,按机构持股高低分组的检验结果表明,机构持股最高的组合,其AQfactor系数的显著程度在10%水平上高于机构持股最低的样本。而在四因子模型中,回归显示并不具有统计意义上的显著性。表明,中国A股投资者在对股票定价时,虽对AQ带来的盈余信息风险有所识别,但无充分的证据表明,AQ能作为一个较为稳定可靠的定价因子来解释股票回报率。
[Abstract]:By using Panel data (fixed effect) measurement method, this paper systematically analyzes the internal factors affecting accrual quality (AQ), and emphatically analyzes the correlation between institutional shareholding and AQ. From the total sample and graded sample, there is a significant negative correlation between AQ and institutional shareholding ratio, which indicates that, on the whole, institutional investors have identified and avoided the stocks of those companies whose earnings fluctuate greatly and AQ is poor. By constructing the alternative variable AQFactor of AQ, it is incorporated into the three-factor and four-factor models, and the time series OLS regression method is used to test the asset pricing. The results show that the AQfactor is positively correlated with the return on stock, but in the three-factor model, The regression coefficient is only significant at the level of 10%. At the same time, the test results according to the level of institutional shareholding show that the combination with the highest institutional shareholding, The significance of AQfactor coefficient in 10% level is higher than that in the sample with the lowest institutional shareholding. But in the four-factor model, the regression shows no statistical significance. Although the risk of earnings information brought by AQ is recognized, there is no sufficient evidence that AQ can be used as a stable and reliable pricing factor to explain the stock return.
【作者单位】：西南交通大学经济管理学院;
【基金】：国家自然科学基金资助项目(70971110) 教育部长江学者和创新团队发展计划资助项目(IRT0860)
【分类号】：F224;F832.51

【参考文献】