Can Artificial Intelligence Trade the Stock Market?
This addresses the problem of improving automated trading strategies for investors, though it is incremental as it applies existing DRL methods to financial data.
The paper tackled stock market trading by applying Deep Reinforcement Learning (DDL) algorithms like DDQN and PPO, showing they outperform a Buy and Hold benchmark and supervised learning methods in risk-adjusted returns across assets like currency pairs, S&P 500, and Bitcoin from 2019-2023.
The paper explores the use of Deep Reinforcement Learning (DRL) in stock market trading, focusing on two algorithms: Double Deep Q-Network (DDQN) and Proximal Policy Optimization (PPO) and compares them with Buy and Hold benchmark. It evaluates these algorithms across three currency pairs, the S&P 500 index and Bitcoin, on the daily data in the period of 2019-2023. The results demonstrate DRL's effectiveness in trading and its ability to manage risk by strategically avoiding trades in unfavorable conditions, providing a substantial edge over classical approaches, based on supervised learning in terms of risk-adjusted returns.