arXiv q-fin· Yuki Kinoshita, Yumiharu Nakano·· 5 hr agoAI Score31
Kernel-Based Collocation Methods for the Heath-Jarrow-Morton Model with Musiela Parameterization
Kernel-based collocation methods for Heath-Jarrow-Morton models with Musiela parametrization
AI summary
This paper studies a kernel-based collocation numerical solution method for the Heath-Jarrow-Morton model under Musiela parameterization. The method is an Euler-Maruyama approximation of a finite-dimensional stochastic differential equation, which can be integrated with standard Monte Carlo workflows for derivative pricing. Conditional mean squared error estimates are derived under explicit stability assumptions, and the method's effectiveness is validated through two finite-resolution examples: the Vasicek model and caplet pricing.
Source:arXiv q-fin · arxiv.org