Adaptive Portfolio Optimization with Multi-Agent Deep Reinforcement Learning and Short-Term Performance Analysis

This research presents a novel portfolio optimization framework using deep reinforcement learning (DRL). Traditional methods rely on static models or single-agent strategies, which struggle with market dynamics. We propose a dynamic system to address this by selecting the best-performing DRL agent based on recent market conditions. The framework evaluates five DRL agents, A۲C, SAC, TD۳, DDPG, and PPO, allocating portfolio weights based on short- term performance. A selection mechanism identifies the top agent using cumulative returns over the prior ten days, leveraging multiple agents' strengths. This adaptive approach embraces the philosophy that no single strategy consistently outperforms in all market conditions, making flexibility and continuous learning essential for robust portfolio management. Backtesting on Dow Jones data shows our method enhances cumulative returns and risk- adjusted performance, achieving an ۱۱.۴۳% average annual return, ۳۸.۲۹% cumulative returns, and a ۰.۸۳۲ Sharpe ratio, outperforming individual DRL agents.

کلیدواژه ها:

Deep Reinforcement Learning ، Consulting System ، Portfolio Optimization ، Tactical Decision ، Cumulative Return