Portfolio Optimization under Recursive Utility via Reinforcement Learning
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908909291175936 |
|---|---|
| author | Chang, Minkey |
| author_facet | Chang, Minkey |
| contents | We study whether a risk-sensitive objective from asset-pricing theory -- recursive utility -- improves reinforcement learning for portfolio allocation. The Bellman equation under recursive utility involves a certainty equivalent (CE) of future value that has no closed form under observed returns; we approximate it by $K$-sample Monte Carlo and train actor-critic (PPO, A2C) on the resulting value target and an approximate advantage estimate (AAE) that generalizes the Bellman residual to multi-step with state-dependent weights. This formulation applies only to critic-based algorithms. On 10 chronological train/test splits of South Korean ETF data, the recursive-utility agent improves on the discounted (naive) baseline in Sharpe ratio, max drawdown, and cumulative return. Derivations, world model and metrics, and full result tables are in the appendices. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_22880 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Portfolio Optimization under Recursive Utility via Reinforcement Learning Chang, Minkey General Finance Computational Engineering, Finance, and Science Portfolio Management We study whether a risk-sensitive objective from asset-pricing theory -- recursive utility -- improves reinforcement learning for portfolio allocation. The Bellman equation under recursive utility involves a certainty equivalent (CE) of future value that has no closed form under observed returns; we approximate it by $K$-sample Monte Carlo and train actor-critic (PPO, A2C) on the resulting value target and an approximate advantage estimate (AAE) that generalizes the Bellman residual to multi-step with state-dependent weights. This formulation applies only to critic-based algorithms. On 10 chronological train/test splits of South Korean ETF data, the recursive-utility agent improves on the discounted (naive) baseline in Sharpe ratio, max drawdown, and cumulative return. Derivations, world model and metrics, and full result tables are in the appendices. |
| title | Portfolio Optimization under Recursive Utility via Reinforcement Learning |
| topic | General Finance Computational Engineering, Finance, and Science Portfolio Management |
| url | https://arxiv.org/abs/2603.22880 |