Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Evaluating the Predictive Power of Dictionary News Sentiment and Commercially Provided Sentiment on Equity Volatility |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 118 |
| Date | 2023 |
| Zusammenfassung | The aim of this thesis is to investigate the effectiveness of the news headline sentiment of a financespecific dictionary, the Loughran-McDonald Master Dictionary, in forecasting the next day’s volatility and downside deviation. The dictionary contains negative and positive words, and the sentiment is generated using a bag-of-words approach. The source of the news headlines is the commercial news analysis database TRNA. The news headlines get filtered by topic codes, which are both companyspecific and of macroeconomic nature. The proprietary sentiment included in the TRNA database is additionally used as a comparison to the Loughran-McDonald sentiment. To better analyze trends in the sentiment, the data gets aggregated to daily levels and smoothed using a crossing exponential weighted moving average. The resulting smoothed sentiments changes are analyzed by performing an ordinary least square regression, where the lagged sentiment value gets regressed on the risk measures. These regressions show that for most topic codes, lagged volatility is significant for both, the TRNA and the Loughran-McDonald sentiment measures. However, the significance is overall higher for the Loughran-McDonald sentiment. |
| Export |
BibTeX
|