Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Deep Learning in Corporate Bonds Pricing |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Date | 2022 |
| Abstract Text | The model and the approach developed by Chen et al. (2019) are replicated and adapted to the framework of corporate bond expected returns. It relies on machine learning methods and estimates the stochastic discount factor for a set of U.S. corporate bonds using macroeconomic, firm-specific and debt-specific data. The thesis finds that accessing corporate bond data over a long period of time can prove challenging but when the negative effect of limited data is ignored, the model seems to perform even better on corporate bond data than on equity data. Furthermore, the thesis studies the predictive power of each type of input. It confirms that the handling of the macroeconomic data suggested by Chen et al. improve the performances. In addition, it shows that debt-specific data have strong predictive power while firm-specific data have predictive power, it does not bring incremental predictive power to the debt-specific data. |
| Export |
BibTeX
|