Efficient pricing of multi-asset options is a challenge in quantitative finance. Fourier methods offer rapid valuation but face issues in high-dimensional settings. The study introduces RQMC as a solution, emphasizing domain transformations for improved convergence rates. Various models and numerical experiments validate the approach's computational advantages over traditional methods.
The Monte Carlo method remains popular but has slow convergence rates, especially for multi-asset option pricing. Fourier methods provide an alternative with faster valuation but struggle in high dimensions due to tensor product structures. The study proposes RQMC as a solution, highlighting domain transformations to enhance convergence rates.
Analytical and numerical smoothing techniques are explored to address challenges in deterministic quadrature methods for option pricing. Fourier methods are presented as efficient alternatives with specific approaches for different models like GBM, VG, and NIG.
The research demonstrates the benefits of applying RQMC in the Fourier space compared to traditional methods like MC or TP-Laguerre quadrature. Domain transformations play a crucial role in improving convergence rates and computational efficiency across various pricing models.
翻譯成其他語言
從原文內容
arxiv.org
深入探究