Multivariate approximation methods for the pricing of catastrophe-linked bonds

Details

Ressource 1Download: BIB_EBF805938DD1.P001.pdf (2367.78 [Ko])
State: Public
Version: author
Serval ID
serval:BIB_EBF805938DD1
Type
Inproceedings: an article in a conference proceedings.
Collection
Publications
Title
Multivariate approximation methods for the pricing of catastrophe-linked bonds
Title of the conference
Modern Developments in Multivariate Approximation: 5th International Conference, Witten-Bommerholz (Germany), September 2002
Author(s)
Albrecher H., Hartinger J., Tichy R.
Publisher
Birkhäuser
Address
Witten-Bommerholz, Germany
ISBN
3764321954
978-3764321956
Publication state
Published
Issued date
2003
Peer-reviewed
Oui
Volume
145
Series
International Series of Numerical Mathematics
Pages
21-39
Language
english
Abstract
In this paper we develop Quasi-Monte Carlo techniques for the evaluation of high-dimensional integrals that occur in financial applications, namely in the pricing of default-risky catastrophe-linked bonds in a model including stochastic interest rates, basis risk and default risk. It is shown that these techniques clearly outperform classical Monte Carlo integration in terms of efficiency. The methods are based on number-theoretic low-discrepancy sequences such as Halton, Sobol and Faure sequences.
Web of science
Create date
12/05/2009 10:56
Last modification date
20/08/2019 16:14
Usage data