Dynamic term structure modeling beyond the paradigm of absolute continuity

Abstract: This thesis is devoted to the study of term structure modeling in interest rate markets and defaultable term structure modeling in credit risk markets.

Post-crisis interest rate markets possess two main characteristics: multiple curves and discontinuities. While a lot of effort has been put in the study of the former, there is one crucial feature of discontinuities, which we will call stochastic. discontinuities, whose investigation seems to be lacking in the interest rate literature so far. This concept of discontinuities has recently been studied in a credit risk framework in Fontana and Schmidt (2018) and Gehmlich and Schmidt (2018). Stochastic discontinuities describe jumps in the underlying interest rates or processes depicting events occurring at announced dates but with a possibly unanticipated outcome. This type of events is clearly present in interest rates, as can be evidenced by jumps in the underlying rates in correspondence with meetings of the European Central Bank. We provide a general analysis of the term structure modeling of multiple curves with the presence of stochastic discontinuities and derive conditions to ensure absence of arbitrage. In particular, we provide an extended Heath-Jarrow-Morton formulation with semimartingales as driving processes. Beyond that, a general market model approach is investigated and some insightful examples in an affine framework are presented in order to show the potential of this approach. Bond prices are calibrated in a Vasi cek framework by means of machine learning techniques adapted to Gaussian processes.

In credit risk we are concerned with securities that are subject to default risk. We present a general analysis of the term structure modeling of defaultable bonds allowing for discontinuities. In particular, we derive conditions to ensure absence of arbitrage in the credit risky financial market in an extended Heath-Jarrow-Morton framework with semimartingales as driving processes. We provide a similar characterization for defaultable bonds with recovery

Objekt anzeigen

• Standort: Deutsche Nationalbibliothek Frankfurt am Main
• Umfang: Online-Ressource
• Sprache: Englisch
• Anmerkungen: Universität Freiburg, Dissertation, 2019

Klassifikation und Themen

• Klassifikation: Naturwissenschaften
• Schlagwort: Zinsstruktur ; HJM-Modell ; Semimartingal ; Arbitrage ; Ausfallrisiko

Beteiligte, Orts- und Zeitangaben

• Ereignis: Veröffentlichung
• (wo): Freiburg
• (wer): Universität
• (wann): 2019
• Urheber: Gümbel, Sandrine
• Beteiligte Personen und Organisationen: Schmidt, Thorsten ; Albert-Ludwigs-Universität Freiburg. Fakultät für Mathematik und Physik

Weitere Informationen

• DOI: 10.6094/UNIFR/151675
• URN: urn:nbn:de:bsz:25-freidok-1516750
• Rechteinformation: Der Zugriff auf das Objekt ist unbeschränkt möglich.