我国创业板IPO首日高频量价分位相关的变点分析

发布时间：2019-05-18 15:01

【摘要】：提出含有结构变点的面板数据分位数回归模型,给出基于贝叶斯推理和MCMC模拟的参数估计方法,对我国创业板IPO上市首日的收益率与交易量、大盘走势之间的日内高频变结构相关性进行实证研究发现:在中高收益率分位点上结构变点均出现在开盘阶段附近,而在较低的分位点上变点位置向后推延至中盘附近;IPO交易量对收益率在变点之前的影响要大于在变点后的影响,而深证成分指数交易量对收益率在变点后的影响要大于在变点前的影响;IPO交易量对收益率的影响在变点前的较低分位点、变点后的任意分位点上均呈现出"量利分离"特殊现象;深证成分指数交易量与新股收益率在变点前后任意分位点上均呈现正相关,但影响系数呈现变点后大于变点前、极端分位大于中间分位的特征.
[Abstract]:In this paper, a quantile regression model of panel data with structural change points is proposed, and a parameter estimation method based on Bayesian reasoning and MCMC simulation is given to determine the rate of return and trading volume on the first day of IPO listing on gem in China. The intra-day high frequency variation structure correlation between the market trends is empirically studied. It is found that the structural change points appear near the opening stage at the middle and high yield sub-points, while the position of the change points at the lower sub-points is postponed back to near the middle market. The influence of IPO trading volume on the rate of return before the change point is greater than that after the change point, while the influence of the trading volume of the Shenzhen stock index on the rate of return after the change point is greater than that before the change point. The effect of IPO trading volume on the rate of return shows a special phenomenon of "separation of quantity and profit" at any sub-point before the change point and at any sub-point after the change point. There is a positive correlation between the trading volume of Shenzhen Stock Index and the yield of new shares at any sub-point before and after the change point, but the influence coefficient is larger than that before the change point, and the extreme score is larger than the middle score.
【作者单位】：中国矿业大学管理学院;
【基金】：国家自然科学基金(71071153) 教育部新世纪优秀人才支持计划(NCET-12-0955) 江苏省333高层次人才培养工程专项资助
【分类号】：F832.51;F224

【参考文献】