首页 学院概况 学位教育 高层培训 师资团队 科学研究 国际交流 党建之窗 学生工作 实验中心 情铸经管 English Version 下载专区
新闻通知
 
 学院新闻
 通知公告
 学术报告
 
  您的位置:首页 > 新闻通知 > 学术报告
柏林洪堡大学陈怡璇教授学术讲座
作者:刘映池    发布时间:2017-09-13 20:21:57    访问次数:3439  
 

讲座时间:2017年9月21日 下午2:30

讲座地点:信息楼0215#

主讲人简介:陈怡璇,柏林洪堡大学商业与经济学院任教轨道教授,及国际科研训练组1792 - 高维非固定时间序列教授。她的研究兴趣集中在“金融和会计分析的文本挖掘”和“风险建模与管理”两个方面。她最近致力于文字挖掘技术的开发和应用。使用这些技术,她量化社交媒体和纳斯达克新闻的情绪。应用统计分析(如机器学习(例如SVM),词典投影,潜在语义分析,Latent Dirichlet分配和主题建模),(1)分析新闻对预期的企业价值之影响; (2)从10-K申报中检测财务报表欺诈; (3)衡量会计报表中与收益预测有关的语言内容。她已在重点期刊上发表,并撰写了金融计量经济学的重要软件。她具有银行业风险建模与管理方面的专业经验。她目前正在洪堡大学和德意志银行之间进行“产学合作计划”,着重信用风险模型和压力测试。

讲座题目:Textual Sentiment, option information and stock Predictability(文本情感、期权信息与股票收益可预测性)

讲座内容简介:

A growing literature shows a predictability of stock returns based on sentiment proxies.More recently, it has been shown that also variables implied from single stock options markets carry predictive content for future equity returns. Where does this predictability stem from? Is it firm-specific information advantage or is it a firm-specific sentiment that is implemented in terms of option-based strategies and thus leads to return predictability?

In this work, we aim at answering this question. We distill sentiment from a huge bulk of NASDAQ news articles and examine the various sources of predictive power. We find that options markets react to sentiment from NASDAQ articles in that higher implied volatility, higher out-of-money put prices and stronger smirk can be observed as more negative articles being posted which constitutes more negative sentiment. Next we inspect return predictability. We find that options variables indeed predict stock returns, yet sentiment variables, in particular, our index sentiment remains a highly relevant factor for individual stock returns. Firm specific-sentiment becomes weaker after controlling for information implied in options. The strength of predictions appears to subside from high to low attention firms, but still remains for low-attention firms. We conclude that the predictability of options markets cannot exclusively be attributed to information asymmetry but also to sentiment.

欢迎各位老师、同学到场交流!

 
学院办公地址:信息楼0606 | 电话:028-87600822/87601403 | 邮编:610031
Copyright © 2005 经济管理学院 All rights reserved 设计与技术支持:经管院信息中心
本站内容未经我院授权,任何单位及个人可下载但不得转载 管理登录   旧版网站