Gradient boosting-based numerical methods for high-dimensional backward stochastic differential equations

作者：

Highlights：

• A novel boosting-based algorithm is proposed for numerically solving nonlinear high-dimensional backward stochastic differential equations (BSDEs) and PDEs.

• A rigorous analysis of the convergence and time complexity is provided.

• Several numerical examples with very high space dimensions are provided to demonstrate the accuracy and performance.

摘要

•A novel boosting-based algorithm is proposed for numerically solving nonlinear high-dimensional backward stochastic differential equations (BSDEs) and PDEs.•A rigorous analysis of the convergence and time complexity is provided.•Several numerical examples with very high space dimensions are provided to demonstrate the accuracy and performance.

论文关键词：Backward stochastic differential equations (BSDEs),XGBoost,High-dimensional problem,Regression

论文评审过程：Received 16 September 2021, Revised 11 March 2022, Accepted 22 March 2022, Available online 4 April 2022, Version of Record 4 April 2022.

论文官网地址：https://doi.org/10.1016/j.amc.2022.127119