连续指令驱动市场的信息交易概率:一种新的方法  

The probability of informed trading in continuous order driven market:A new method

作　　者：李广川 刘善存[1] 邱菀华[1] 

机构地区：[1]北京航空航天大学经济管理学院,北京100193 [2]深圳证券交易所,深圳518010

出　　处：《管理科学学报》2010年第10期8-20,共13页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(7087100270671006);全国优秀博士学位论文作者专项基金资助项目(200466)

摘　　要：通过构造"成交积极性"变量,提出了一种在连续竞价指令驱动市场中估计信息交易概率的新方法,运用有序probit方法和马尔科夫转换技术对模型进行了估计.对该方法与基于PIN框架的经典信息交易概率模型进行比较分析,同时研究了信息交易概率和资产收益波动、报价价差、成交持续期之关系的日内效应.研究结果发现,模型较之基于PIN框架的经典信息交易概率模型对价差具有更好的解释力;信息交易概率和资产收益波动、报价价差、成交持续期等股票交易特征之关系具有显著的日内效应.该结论经过面板回归模型检验具有稳健性.Based on the variable ＂Transaction Aggressiveness＂ put forward in this paper,a new method to estimate probability of informed trading in continuous auction order driven market is proposed.We then do a comparative analysis between our method and classic models based on framework.We also research on intraday effects of the relations between probability of informed trading and trading characteristics such as asset return volatility,quoted spread and trading duration.We conclude that,firstly,our model has more explanatory power on spread in comparison with classic models based on framework.Secondly,intraday effects of the relations between probability of informed trading and trading characteristics do exist,which are robust based on panel regression analysis.

关 键 词：指令驱动连续竞价市场 信息交易概率 交易特征 日内效应 

分 类 号：F830.9[经济管理—金融学]