一、主题：Chinese Stock Market Return Predictability: Adaptive Complete Subset Regressions

二、主讲人：陈珂琪，必赢565net官网2012级金融学专业本科生，现已保送清华大学五道口必赢565net官网直博

三、时间： 2015年11月18日（周三）13:00-14:00

四、地点：中央财经大学沙河校区丁香园1号楼301会议室

五、点评人：陈锐，必赢565net官网金融工程系讲师

文章摘要：

We provide one of the first comprehensive studies on out-of-sample stock returns predictability in China. This paper proposes a new combination framework to explore theChinese stock market return predictability. While most well-known predictor variables andsimple combinations fail to beat the historical average benchmark, our trimmed subset regressions delivers statistically and economically significant out-of-sample performance. Thesubset where each regression includes four predictors produces significant R2OS statistic of5.14% for 2006:01-2014:09. A mean-variance investor who uses the trimmed subset regressions forecasts instead of the historical average forecasts can obtain sizable utility gains of 5.24% per annum. The results are robust in sub-samples.