【摘要】 检验中国证券交易所四类股票收益与波动的时间序列特征,以及收益与波动之间的关系.首先应用广义自回归条件异方差模型(GARCH)和指数GARCH模型以获得合适的条件方差序列.应用结果发现,波动性随时间变化的证据,并且波动高/低的时期趋向于聚集,显示出高度持续性和可预测性.然后,应用均值GARCH(GARCH-M)模型检验预期收益与预期风险的关系.研究结果认为,每日交易量作为每日信息到达时刻的代理变量对于中国股票市场每日收益的条件波动的解释力度不显著.更多还原

【Abstract】 We examine time series features of stock returns and volatility, as well as the relation between returns and volatility in China’s stock markets. Firstly, GARCH and EG ARCH models are employed to generate conditional variance series. The application of the two models provides strong evidences of time-varying volatility. In addition, the results show that the time of high and low volatility tends to converge and volatility is highly persistent and predictable. Then we test the relation between expected returns and expected risk with the GARCH-M model. We find that daily trading volume, used as a proxy for information arrival time, has no significant explanatory power on the conditional volatility of daily returns.更多还原