论文部分内容阅读
以证监会查处的17只公告前存在内幕交易或信息泄漏的股票为样本,分别利用EKOP模型和DY模型,从交易量和公共信息两个视角,研究A k tas等(2007)发现的信息性交易概率P IN在公告前降低在公告后升高的异常现象。结果表明,交易量和公共信息因素对P IN的估计有重要影响,考虑到这两个因素后,P IN在公告前降低的异常现象消失,但是P IN在公告后相对于公告前上升的异常继续存在。本文认为这是由于公共信息造成的指令流不平衡所致,并指出P IN作为总的买卖指令不平衡性的度量包含了比私人信息更多的内容,高估了市场的信息不对称程度。此外,本文提出了一个新的算法,彻底解决了用计算机对EKOP模型的似然函数进行估计时经常遇到的数值溢出问题。
Using the stocks with insider trading or information leakage before the 17 bulletins investigated by the SFC as a sample, this paper uses the EKOP model and the DY model respectively to study the informational information discovered by A tas (2007) from two perspectives: transaction volume and public information The trading probability P IN reduces the anomaly that rises after the announcement before the announcement. The results show that the transaction volume and public information factors have an important impact on the P IN estimation. Taking into account these two factors, the anomaly that P IN decreases before the announcement disappears. However, the P IN after the announcement relative to the antecedent before the announcement Continue to exist. This paper argues that this is due to the imbalance of instruction flows caused by public information and points out that the measure of P IN as a general trade instruction imbalance contains more content than private information and overestimates the degree of information asymmetry in the market. In addition, a new algorithm is proposed in this paper to completely solve the numerical overflow problem often encountered when using computer to estimate the likelihood function of EKOP model.