论文部分内容阅读
本文基于Log-ACD模型和一类非参数模型,研究了股票价格持续上升时期和价格持续下降时期,交易量久期与价格变化的动态关系。研究表明在不同的市场格局下,价格变化对交易量的影响会有显著区别。另一方面我们发现阈值的选取会影响交易量久期的统计性质,阈值变大时交易量久期的长记忆性会变弱。本文在理论上也有所创新,采用了本文前两位作者提出的新的方法估计Log-ACD模型的参数,该方法在误差服从厚尾分布时具有良好的统计性质。利用新的估计构造了Wald检验统计量,检验价格变化的方向对预期交易量久期是否有显著影响。
Based on the Log-ACD model and a class of nonparametric models, this paper studies the dynamic relationship between the stock price rising period and the price declining period, the trading volume duration and the price change. Research shows that under different market conditions, the impact of price changes on trading volume will be significantly different. On the other hand, we find that the selection of the threshold will affect the statistical properties of the trading volume duration, and the longer the threshold, the longer the memory of the trading volume will be weakened. This paper is also theoretically innovative, using the new method proposed by the first two authors of this paper to estimate the parameters of Log-ACD model. This method has good statistical properties when the error obeys the thick tail distribution. The Wald test statistic is constructed by using the new estimation to test whether the direction of the price change has a significant impact on the expected trading duration.