Comparison Principle and Semi-Lagrangian Methods for a Hamilton--Jacobi--Bellman System with Recursive Utility
math.OC, cs.NA, math.NA
Submitted: 2026-11-02
Updated: 2026-11-02
License: http://creativecommons.org/licenses/by/4.0/
The gist: We consider a system of coupled first-order Hamilton--Jacobi--Bellman equations arising in continuous-time heterogeneous agent models with recursive utility…
Abstract
We consider a system of coupled first-order Hamilton--Jacobi--Bellman equations arising in continuous-time heterogeneous agent models with recursive utility, a class of models that has attracted increasing attention in macroeconomics. The first part of the paper establishes a comparison principle for the system when preferences exhibit early resolution of uncertainty. A key ingredient in the proof is an ordinally equivalent transformation of the recursive utility representation. In the second part, we develop semi-Lagrangian schemes and iterative algorithms for solving the associated nonlinear discrete systems. For preferences exhibiting late resolution of uncertainty, we establish a first-order error estimate for the semi-Lagrangian approximation. Numerical experiments demonstrate that the proposed methods yield results consistent with those obtained using the previously developed finite difference schemes.
Sources
- Mastering Stochastic OLG Models in Continuous Time
- Simulating numerically the Krusell-Smith model with neural networks
- Finite difference methods for a continuous-time heterogeneous agent model with recursive utility
- Global Solutions to Master Equations for Continuous Time Heterogeneous Agent Macroeconomic Models
- Structural Reinforcement Learning for Heterogeneous Agent Macroeconomics
Related papers
- An Accelerated Proximal Gradient Method for Bures-Wasserstein Space
- FOMTools: Development and AI-Assisted Code Generation for First-Order Methods
- Lions and Muons: Optimization via Stochastic Frank-Wolfe under Heavy-Tailed Noise
- Dynamic Service Recommendation with Congestion-Dependent Joining: Near-Optimal and Constant-Factor Approximation Algorithms
- Integrating Learning and Optimization for Open Transport Capacity Pooling under Disruption
- Zeroth-Order Gradient Estimators under Decision-Dependent Noise