Not logged in.

Contribution Details

Type Master's Thesis
Scope Discipline-based scholarship
Title Ahead or behind the curve The impact of public and private information in option-implied sentiment
Organization Unit
Authors
  • Santiago Walliser
Supervisors
  • Matthias Uhl
Language
  • English
Institution University of Zurich
Faculty Faculty of Business, Economics and Informatics
Number of Pages 61
Date July 2021
Abstract Text We investigate the stock return predictive power at the weekly level of an option-implied sentiment time series that captures private information, in combination with public company-specific and market news sentiment. We provide strong evidence that sentiment variables explain variations in option price summa-rizing features at the weekly level. The effects seem to be asymmetric along the moneyness axis. Further-more, we document that aggregate market sentiment has a higher impact than company level sentiment. In addition, we find that news published during the trading period tend to have less information content and are mostly related to the overall stock market. Conversely, news published after the market closes have more impact as they are more complex and thus contain more information content. Using an orthogonal-ization procedure in which we remove the market sentiment drift component in company level sentiments, we are able to increase their information content. Based on these findings, we develop a model to predict stock returns using private and public information extracted from the options market. We discover that ATM implied volatilities orthogonalized to news sentiment have strong predictive power alongside market level consensus. A trading strategy based on this model exhibits a significant alpha with an annualized return of 18.38% and an information ratio of 0.98 over the entire period of almost twenty years analyzed. Key words: stock return predictability, option markets, private information, news sentiment, non-trading hour information, market level sentiment, company-specific sentiment
Export BibTeX