股票收益反转效应及与买卖价差关系研究  

Research on reversal effect and the relationship between reversal effect and bid-ask spread

作　　者：董晨昱[1,2] 刘维奇[2,3] LIU Wei-min 王钰[4] 

机构地区：[1]山西大学数学科学学院,太原030006 [2]山西大学管理与决策研究中心,太原030006 [3]山西财经大学财政金融学院,太原030006 [4]山西大学软件学院,太原030006

出　　处：《管理科学学报》2016年第6期171-182,共12页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(71371113);教育部人文社科基金资助项目(13YJA790154)

摘　　要：研究了买卖价差与股票短期收益反转的关系.拓展了"真实"价格保持不变或遵循随机游走的假设,在三种情形下对比连续时间上观测价格和"真实"价格的路径图及其价格变化的联合分布来分析买卖价差对股票短期收益反转的影响,结果表明,当"真实"收益的一阶自相关系数接近于0时,买卖价差是造成观测收益反转的唯一原因;当"真实"收益有较强的一阶负自相关性时,买卖价差对观测收益反转的作用不再明显,甚至可能减弱此效应;买卖价差加剧了观测收益的波动.通过买卖价差与股票观测收益间的恒等关系,建立了收益分解模型,将"真实"收益从观测收益中分离,随后使用NASDAQ市场的个股日、周和月数据进行横截面回归和方差比检验,实证结果也支持上述结论.The paper analyzes the relationship between the bid-ask spread and stock short-run return reversals in individual stocks. This paper extends the assumption that the ＂true＂ price remains constant or follows random walk. Contrasted paths of observed price and ＂true＂ price between successive time periods, and adjacent period joint distribution of observed price change and the ＂true＂ price change in three cases, we analyze the impact of the bid-ask bounce on the return reversal. The results show, the bid-ask spread is the only reason for stock observed return reversal when the first order autocorrelation coefficient of the stock ＂true＂ return is close to 0 ; the role of the bid-ask spread on observed return reverse is not obvious and may even weaken the reverse when the ＂true＂ return has a strong negative first order autocorrelation; the bid-ask spread intensify the volatility of observed return. Based on the identity relation between bid-ask spread and stock return, we establish the return decomposition model to isolate the ＂true＂ return from the observed return, and cross sec tion regression and variance-ratio test are adopted to support these conclusions using individual stocks daily, weekly and monthly data in NASDAQ market respectively.

关 键 词：收益反转 买卖价差 方差比检验 

分 类 号：F830[经济管理—金融学]