Metraipolis › archive › dc_45285cfce5e8
Simulation-Based Neural Policies for Portfolio Choice: Architecture, Training, and Interpretability — arXiv:2608.03933
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.03933
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:48:34.621Z
- SHA-256 of the archived text
5d0dd933ce0c37322d0c0b8ea0819a0a833772849e7c90b299b2e23420d1fe1c- Stable URL of this page
https://metraipolis.com/documento/dc_45285cfce5e8
How to cite
Simulation-Based Neural Policies for Portfolio Choice: Architecture, Training, and Interpretability — arXiv:2608.03933 — arXiv (en), https://arxiv.org/abs/2608.03933 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_45285cfce5e8 (Metraipolis archive dc_45285cfce5e8, sha256 5d0dd933ce0c37322d0c0b8ea0819a0a833772849e7c90b299b2e23420d1fe1c)
BibTeX
@misc{dc45285cfce5e8,
title = {Simulation-Based Neural Policies for Portfolio Choice: Architecture, Training, and Interpretability — arXiv:2608.03933},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.03933},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_45285cfce5e8; sha256 5d0dd933ce0c37322d0c0b8ea0819a0a833772849e7c90b299b2e23420d1fe1c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03933. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Many economic decision problems, lifecycle consumption-saving and dynamic portfolio choice, are finite-horizon stochastic control problems with continuous states and actions. When the state is low-dimensional these problems are solved by dynamic programming on a grid. The grid cost grows exponentially in the state dimension, known as the curse of dimensionality, which motivates replacing the value-function grid with a neural policy optimized directly through simulation. Such policies are usually studied in the high-dimensional settings that motivate them, precisely where no reference solution exists. So the contribution of any single architectural or training choice cannot be isolated and diagnosed. We therefore take a step back and treat both the architecture and the solution method as the objects of study. To this end, we consider a lifecycle problem with a sufficiently low-dimensional normalized state space to admit an accurate dynamic programming solution, which is used for evaluation. We compare four architectures. The simplest consists of a single time-conditioned network. We then consider two networks concatenated across the regime switch, followed by one network per date trained backward against frozen downstream policies. Finally, we evaluate a constrained variant of the per-date architecture. Decoupling the policy across time gives each date a short, well-posed objective, which we pair with direction-dominant optimization that normalizes away gradient magnitude. Architectures that lead to similar realized utility objective can nevertheless differ in whether they respect the underlying problem's economics. We therefore evaluate each design jointly based on welfare, a solution-free Bellman residual, shape restrictions, and the resulting policy functions. Tema degli esploratori: matematica e informatica teorica (retro-catalogo arXiv: annunci del 2026-08-05 in math.OC). Categorie arXiv: Optimization and Control (math.OC). Pagina del lavoro: https://arxiv.org/abs/2608.03933 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T12:35:48.818Z dalla pagina https://arxiv.org/catchup/math.OC/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).