Data-Driven Dynamic Portfolio Selection with Vine Copula Simulation and Reinforcement Learning

By Gabriël M. J. Gelissen, Fengyi Yuan

Rating

1817
Battle Count: 50

Relevance

9/10
Highly relevant. The paper addresses dynamic portfolio selection, a core problem in quantitative trading, by integrating advanced dependence modeling (Vine Copulas) with modern Deep Reinforcement Learning (TD3/LSTM). It provides both theoretical stability bounds and empirical evidence of improved risk-adjusted returns and reduced turnover compared to benchmarks.

Implementation Complexity

8/10
High complexity. Requires implementing a neural vine copula model (including marginal GARCH/Student-t fits and dynamic pair-copula networks), a recurrent TD3 agent with LSTM, and a constrained action mapping layer. The integration of simulation-based CVaR evaluation within the RL loop adds significant computational overhead.

Reproducibility

3/5
The paper provides detailed numerical configurations (Appendix A), algorithm steps (Algorithm 1), and mathematical formulations. Data sources are identified (Sina Finance via AKShare). However, no public code repository URL is explicitly provided in the text.

Abstract

We develop a data-driven approach to dynamic portfolio selection that combines vine-copula simulation with reinforcement learning. The investor has constant relative risk aversion and maximises expected terminal utility with a running penalty on the conditional value-at-risk (CVaR) of one-period losses. We estimate a conditional vine model from historical returns and use it to generate joint scenarios for CVaR evaluation and conditional paths for training a recurrent portfolio policy. Under uniform local derivative bounds, we propagate edgewise pair-copula parameter errors through the conditional transforms to obtain power-law stability bounds in Hellinger distance and total variation. In a finite-horizon investment model, we combine these estimates with a marginal and structural approximation budget to bound policy values, optimal values and the true-market loss of policies selected under a fitted model, with separate simulation and numerical optimisation terms. On a seven-asset panel comprising US and Chinese equities and gold, the selected ensemble of ten trained policies achieves a compound annual growth rate of 34.13% and a Sharpe ratio of 2.31 over 22 complete monthly holding periods, exceeding the corresponding values of six classical and vine-based benchmarks. Combining simulated and historical training data reduces monthly target-to-target turnover by 41.5% relative to training on historical data alone, while maintaining comparable growth.

Links

arXiv:2610.20037 · Paper rankings

The interactive Everscope explorer (charts, battles, favorites) loads below.