中文摘要：

本文从个股横截面数据提取尾部风险作为时变罕见灾难风险的代理指标,实证分析了罕见灾难风险作为定价因子对我国股市收益的预测能力和横截面收益的解释能力.基于中国A股市场1997-2013年横截面日收益率数据,利用极值理论和尾部幂指数分布统计量,提取动态尾部风险指标,实证研究表明：1）尾部风险对我国一个月、半年、一年和两年股市收益率具有显著的预测效果;2）尾部风险对于我国股市横截面收益也具有显著的解释能力,尾部风险因子载荷系数大的投资组合所获得的风险补偿显著大于尾部风险因子载荷系数小的投资组合.此外,实证分析还发现尾部风险同宏观经济变量有着显著地反向关系,这进一步验证了尾部风险作为罕见灾难风险度量的合理性.

英文摘要：

From the daily individual stock returns during 1997-2013 in China, we construct a time-varying tail risk measure as a proxy for the rare disaster risk. We empirically find that： 1） the tail risk measure has a significant predictability power for one-month, half-year, one-year and two-year ahead moving average market returns; 2） the tail risk measure can also explain the cross sectional difference of returns, the higher tail risk loading portfolios require a relatively higher risk premium compared to the lower tail risk loading portfolios. Furthermore, we find that there is negative relationship between tail risk and aggregate Maeroeeonomic variables, which further justify the tail risk as a proxy for the disaster risk measure.