Deep Q-Network Trading Trained on Synthetic Market Trajectories

Summary

The paper introduces Trading Deep Q-Network (TDQN), a deep reinforcement learning approach for choosing stock market positions over time. It adapts the DQN method to trading and sets the Sharpe ratio as the performance measure the strategy seeks to maximize. The described objective is to find suitable positions during trading, rather than simply apply an unchanged general-purpose reinforcement learning algorithm.

The agent is trained using artificial trajectories generated from a limited set of historical stock data. The paper also proposes a more rigorous method for evaluating trading strategies and reports promising TDQN results under that assessment. The provided description does not specify the markets, dataset details, benchmark strategies, numerical results, or evaluation procedures, so it is not possible to judge the reported performance or its generalizability from this text alone. Synthetic trajectories and limited historical data are central features of the approach, but their construction and implications are not explained here.

Key ideas

• TDQN adapts deep Q-learning to choose trading positions in stock markets.
• The strategy is designed to maximize Sharpe ratio performance.
• Training uses artificial trajectories generated from a limited historical dataset.
• The paper proposes a more rigorous performance assessment method and reports promising results.
• The available description omits implementation details and numerical evidence needed to assess robustness.

Tags

# An Application of Deep Reinforcement Learning to Algorithmic Trading


# An Application of Deep Reinforcement Learning to Algorithmic Trading









This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in stock markets. It proposes a novel DRL trading strategy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new trading strategy is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN strategy.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.