Asian Option Pricing Using the Physics-Informed Neural Networks Method

초록

Accurately calculating the prices ofAsian options is challenging due to their path- dependent characteristics and the high-dimensional nature of the problem. This study addresses this issue using a novel Physics-Informed Neural Network (PINN) approach, which leverages the strengths of both neural networks and partial differential equation methods. By applying this PINN method to the pricing problems of one-asset and two-asset Asian options, we demonstrate that it can efficiently produce accurate price estimates compared to the traditional Monte Carlo method.

키워드