Modeling and analysis of structured finance products [thesis]

Florian Kramer, Universität Ulm
Structured finance products represent a relatively young asset class with the first deals dating back to the 1970s. The present thesis provides a framework for modeling such securities. After a detailed discussion of its properties, we apply this framework in order to model and empirically analyze two different classes of structured finance securities: structured credit products and mortality contingent catastrophe bonds. Regarding the question of how to explain the risk characteristics of
more » ... tured credit products, we analyze in detail the relevance of contagion effects for modeling the default clustering in the underlying portfolio. In particular, we find that in practical model applications estimation errors are by far more influential than deviations stemming from the assumption of conditional independence. For modeling mortality contingent catastrophe bonds, on the other hand, we show that an appropriate model should consist of at least two components: one relating to regular fluctuations of mortality rates and another one driven by jumps, where the latter represents catastrophic events. However, it is necessary to point out that the catastrophe component will always be subject to high parameter uncertainty since data on past events is sparse.
doi:10.18725/oparu-1084 fatcat:t3rhl2udknepdikwzmttlm2ama