Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Deep calibration of Financial Market Risk Models |
| Other Titles | Event set generation using robust historic resampling |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 30 |
| Date | 2019 |
| Abstract Text | In this thesis, we will develop an extreme value model for four asset classes, which are stocks, real estate, commodities, and bond spreads. Standard techniques in the academic field use the extreme value theory, while industry typically will only use the last one or two crisis and non-crisis data. But they all have their own issues. This thesis analyzes information from crisis further in the past (e.g. 120 years). It uses this to estimate the severity of potential future crises. With the goal to address the criticism that financial market risk models always underestimate the severity of the next crisis, we will use the peak-over-threshold method to screen the crises and Jackknife to recreate the extreme loss distribution, and finally generate the model with a resampling method. The result of this thesis can be used for capital adequacy risk models for regulatory purposes and internal steering. |
| Export |
BibTeX
|