Methoden-Block I / 11.00-12.30 Uhr
InhaltAbout ten years ago, researchers in business administration and economics started using textual analysis methods to systematically analyze the verbal information of documents like, for example, newspaper articles, firm's financial reports, analyst and credit rating reports, internet message boards and transcripts of central bank meetings.
In the first part of this workshop, participants will be introduced to the popular dictionary-based textual analysis approach, which has been established by Tetlock (2007) and Loughran and McDonald (2011). After a short presentation of the approach and the most commonly used dictionaries, workshop participants will get an introduction to the implementation of the approach in Python. In the second part, alternative methods for measuring a text's sentiment (e.g., Naïve Bayes) will be discussed. The workshop concludes with an outlook on how to quantify further dimensions of texts like, for example, a text's readability and a text's similarity to other texts.
Additional information: As the methods covered in this workshop can be applied to many different settings, this workshop
targets students/researchers from all areas, including accounting, economics, finance, marketing, and management. It is not required to have previous knowledge or experience in Python and/or to be familiar with the topic of textual analysis.
References
Loughran, T., and McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10‐Ks. The Journal of Finance, 66(1), 35-65.
Tetlock, P. C. (2007). Giving content to investor sentiment: The role of media in the stock market. The Journal of Finance, 62(3), 1139-1168.
DozentProf. Dr. Alexander Hillert ist Inhaber der House of Finance-Professur für Sustainable Asset Management an der Goethe-Universität Frankfurt. Seine Forschungsschwerpunkte liegen in den Bereichen Asset Management, Behavioral Finance, Empirical Asset Pricing und Textual Analysis.
