On the Risk Assessment of German Covered Bonds in a One-Period Setting
- This research examines the events of mortgage Pfandbrief defaults occurring due to asset-liability mismatches on the balance sheet of Pfandbrief banks. The risk assessment of the Pfandbrief in a one-period and risk-neutral setting is based on advanced structural and reduced-form modelling approaches. In its over 200 year history not one single Pfandbrief has ever defaulted, however, this practically risk-free perception has changed since the recent financial crises. Investors seek methods to carry out their own credit quality analysis instead of relying on ratings from third parties, e.g. rating agencies, and basing their investment decisions thereupon. A generic modelling framework is proposed and adapted to the mortgage type of Pfandbrief. Certain characteristics attributed specifically to the Pfandbrief and cover pool risks, for example, refinancing risk, interest rate risk and asset default risk are incorporated into the introduced models. The first model is based on Merton’sThis research examines the events of mortgage Pfandbrief defaults occurring due to asset-liability mismatches on the balance sheet of Pfandbrief banks. The risk assessment of the Pfandbrief in a one-period and risk-neutral setting is based on advanced structural and reduced-form modelling approaches. In its over 200 year history not one single Pfandbrief has ever defaulted, however, this practically risk-free perception has changed since the recent financial crises. Investors seek methods to carry out their own credit quality analysis instead of relying on ratings from third parties, e.g. rating agencies, and basing their investment decisions thereupon. A generic modelling framework is proposed and adapted to the mortgage type of Pfandbrief. Certain characteristics attributed specifically to the Pfandbrief and cover pool risks, for example, refinancing risk, interest rate risk and asset default risk are incorporated into the introduced models. The first model is based on Merton’s structural approach. Significant improvements wrt to computation time and accuracy of the underlying stochastic process are accomplished as well as an enhanced least square Monte Carlo application is established. The innovative reduced-form approach adopts techniques from CDO modelling which are applied to the underlying cover pool. A large homogeneous portfolio is postulated for the cover pool where stochastic recovery rates are included. Forecasting future default probabilities in continuous time is accomplished via generator matrices where updated credit quality information regarding the cover pool assets can be integrated. Furthermore, the well-established JLT model is extended allowing stochastic risk premiums to be exogenously considered. Dependency between asset positions is captured via copulas. The profound advantage of the reduced-form model over its structural counterpart is the more accurate modelling of the Pfandbrief’s downside risk.…
Author: | Maximilian Hughes |
---|---|
URN: | urn:nbn:de:bvb:384-opus4-679493 |
Frontdoor URL | https://opus.bibliothek.uni-augsburg.de/opus4/67949 |
Advisor: | Ralf Werner |
Type: | Doctoral Thesis |
Language: | English |
Year of first Publication: | 2020 |
Publishing Institution: | Universität Augsburg |
Granting Institution: | Universität Augsburg, Mathematisch-Naturwissenschaftlich-Technische Fakultät |
Date of final exam: | 2019/11/05 |
Release Date: | 2020/02/10 |
Tag: | Deckungsstock; Fristentransformation; Risikoneutrale Bewertung; Bankaktiva und -passiva; Kredit- und Zinsrisiko credit and interest rate risk; large homogeneous portfolio; stochastic recovery rates and risk premiums; copula; covered bond; cover pool; asset-liability mismatch; one-period setting; risk-neutral valuation |
GND-Keyword: | Wirtschaftsmathematik; Hypothekenbank; Pfandbrief; Risikoanalyse; Kreditrisiko; Zinsänderungsrisiko |
Pagenumber: | 300 |
Institutes: | Mathematisch-Naturwissenschaftlich-Technische Fakultät |
Mathematisch-Naturwissenschaftlich-Technische Fakultät / Institut für Mathematik | |
Mathematisch-Naturwissenschaftlich-Technische Fakultät / Institut für Mathematik / Lehrstuhl für Rechnerorientierte Statistik und Datenanalyse | |
Dewey Decimal Classification: | 5 Naturwissenschaften und Mathematik / 51 Mathematik / 510 Mathematik |
Licence (German): | Deutsches Urheberrecht |