University of Cambridge > > Cambridge Psychometrics Centre Seminars > Textual sentiment, option information and stock predictability

Textual sentiment, option information and stock predictability

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Professor John Rust.

A huge amount of literature show a predictability of options markets for future equity returns, but to which the extent the predictability can be created is a major research question. Is the predictability stemming from information advantage or from investor sentiment for firm perspective? The prevalence of social media platform along with textual analytics enable us to distil sentiment and examine the source of predictive power. We find options markets react to sentiment from NASDAQ news. A higher implied volatility, higher out-of-money put price and a higher smirk can be observed as more negative articles being posted which constitutes more negative sentiment. While excluding the sentiment component, we find the predictability of option variables is very limited for high-attention firms but still remains for low-attention firms. In sum, the predictability of options markets is not only attributed to information asymmetry but also sentiment.

Wolfgang Karl Hardle is Ladislaus von Bortkieviecz Professor of Statistics at Humboldt-Universität zu Berlin and Director of CASE (the Center for Applied Statistics & Economics). He is also Director of the Collaborative Research Center CRC649 “Economic Risk” and of the Sino German International Research Training Group IRTG1792 „High dimensional non stationary time series analysis“ (WISE, Xiamen University). His research focuses on dimension reduction techniques, computational statistics and quantitative finance. He has published 34 books and more than 250 papers in top statistical, econometrics and finance journals. He is one of the “Highly cited Scientist” according to the Institute or Scientific Information. He has professional experience in financial engineering, structured product design and credit risk analysis.

Cathy Yi-Hsuan Chen is a full-time, 2-year tenure-track associate professor at the School of Business & Economics in Humboldt-Universität zu Berlin, and a principal investigator of the International Research Training Group 1792 – High Dimensional Non Stationary Time Series. Her research interests focus on “text mining on finance analysis” and “risk analysis and management”. She has dedicated herself recently on applying text mining techniques to distil news flow from social media. Using the statistical analytics such as Machine Learning (e.g. SVM ), Lexicon Projection, Latent Semantic Analysis, Latent Dirichlet Allocation and Topic Modelling, she analyzes the news impact on financial markets. She has published in key journals and has written important software for financial econometrics. She applies modern econometric techniques, such as copulae and ultrahigh dimensional factor models to financial data on systemic risk indicators. She has professional experience in derivative pricing and trading, risk modeling and management in banking industry. She is currently heading a “transfer project” between Humboldt-Universität and Deutsche Bank, and focusing on credit risk modelling and stress testing.

This talk is part of the Cambridge Psychometrics Centre Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2019, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity