中国A股市场过度波动之谜——基于日个股回报率的实证研究

摘要

本文首次基于方差差（Variance Difference）指标对股票回报率过度波动进行量化,并分别运用组合研究和Fama-Macbeth回归的方法,从过度波动与股票回报率之间关系的角度研究了中国A股市场的过度波动现象。基于1995-2010年A股日回报率的实证结果表明,A股市场日回报率波动中存在投资者过度反应导致的过度波动,过度波动带来的额外风险＂驱逐＂了风险回避的理性投资者,使得股市出现低估。通过每月买入过度波动最大的20%股票,同时卖出过度波动最小的20%股票,能获得显著为正的超额收益,并且该超额收益不能被市场风险所解释。%This article quantifies excess volatility by variance difference for the first time in the literature,and employs portfolio approach and Fama-Macbeth regression to research the puzzle of excess volatility in Chinese A share market.The empirical evidence over the sample period of 1995 to 2010 shows that the excess volatility cased by overreaction does exist in Chinese A share market.Further,such excess volatility brings additional risks for rational investors with risk-averse utility functions.Since risk-averse investors underweight the stocks with excess volatility,these stocks can be undervalued relative to other stocks.Indeed,significant abnormal returns can be obtained by buying the quintile portfolio with the largest excess volatility and selling the quintile portfolio with the least excess volatility every month.