摘要
本研究通过分析机构调研信息文本中的关键词,构建了调研活动信息内容的测度框架,并研究了基于调研日的调研内容月度时序对中国股票收益率的预测能力。研究发现,机构投资者调研活动信息内容在样本内和样本外均能显著预测下月中国股票市场的超额收益率,且相较于经济变量具有更好的预测力。机构投资者调研时所获取的信息内容可作为预测股票收益率的信号并通过影响投资者情绪,从而对股票市场的收益率产生影响。通过文献比较表明,基于机构调研日的调研信息时序相较于基于调研报告披露日的调研信息时序在股票收益预测中更具价值。
This paper constructs the structure of site visit information content measurement through analyzing the keywords of institutional investors’ site visit information, and studies the predictability of site visit information on site visit day on stock market return.The study finds institutional investors’ site visit information content could predict Chinese stock market return in in-sample and out-of-sample, which also has a better predictability than economic variable.The site visit information content on site visit day could be regarded as signal of predicting stock market return.The site visit information contents on site visit day have effects on investors’ sentiment,then it affects stock market return.Compared with exist literaturem the forecasting power on stock market return of site visit information on site visit day is more valuable than site visit information on disclosure day.
作者
岳思诗
曹嘉威
YUE Si-shi;CAO Jia-wei(School of Economic and Management,Southwest Jiaotong University,Chengdu 610031,China)
出处
《系统工程》
CSSCI
北大核心
2020年第6期97-107,共11页
Systems Engineering
基金
国家自然科学基金资助项目(71862003,71773100)。
关键词
文本分析
调研活动信息内容
股票收益率
预测力
Text Analysis
Site Visit Information Content
Stock Market Return
Forecasting Power
