Deep Reinforcement Learning on a Multi-Asset Environment for Trading

  title={Deep Reinforcement Learning on a Multi-Asset Environment for Trading},
  author={Ali Hirsa and Branka Hadji Misheva and Joerg Osterrieder and J. A. Posth},
  journal={International Political Economy: Investment \& Finance eJournal},
Financial trading has been widely analyzed for decades with market participants and academics always looking for advanced methods to improve trading performance. Deep reinforcement learning (DRL), a recently reinvigorated method with significant success in multiple domains, still has to show its benefit in the financial markets. We use a deep Q-network (DQN) to design long-short trading strategies for futures contracts. The state space consists of volatility-normalized daily returns, with… Expand


Financial Trading as a Game: A Deep Reinforcement Learning Approach
An Markov Decision Process (MDP) model suitable for the financial trading task and solve it with the state-of-the-art deep recurrent Q-network (DRQN) algorithm and develops an action augmentation technique to mitigate the need for random exploration by providing extra feedback signals for all actions to the agent. Expand
Deep Reinforcement Learning for Trading
The experiments show that the proposed algorithms can follow large market trends without changing positions and can also scale down, or hold, through consolidation periods, and are equivalent if a linear utility function is used. Expand
Practical Deep Reinforcement Learning Approach for Stock Trading
The proposed deep reinforcement learning approach is shown to outperform the two baselines in terms of both the Sharpe ratio and cumulative returns. Expand
Deep Direct Reinforcement Learning for Financial Signal Representation and Trading
This work introduces a recurrent deep neural network for real-time financial signal representation and trading and proposes a task-aware backpropagation through time method to cope with the gradient vanishing issue in deep training. Expand
A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem
A financial-model-free Reinforcement Learning framework to provide a deep machine learning solution to the portfolio management problem, able to achieve at least 4-fold returns in 50 days. Expand
Enhancing Time Series Momentum Strategies Using Deep Neural Networks
Backtesting on a portfolio of 88 continuous futures contracts, it is demonstrated that the Sharpe-optimised LSTM improved traditional methods by more than two times in the absence of transactions costs, and continue outperforming when considering transaction costs up to 2-3 basis points. Expand
Optimal Asset Allocation using Adaptive Dynamic Programming
Asset allocation is formalized as a Markovian Decision Problem which can be optimized by applying dynamic programming or reinforcement learning based algorithms and is shown to be equivalent to a policy computed by dynamic programming. Expand
The Applicability of Self-Play Algorithms to Trading and Forecasting Financial Markets
This study shows enormous potential of applying self-play concepts and algorithms to financial markets and economic forecasts, and shows that academic research using Self-Play has mostly not focused on finance, and if it has, it was usually restricted to stock markets, not considering the large FX, commodities and bond markets. Expand
Portfolio Management using Reinforcement Learning
In this project, we use deep Q-learning to train a neural network to manage a stock portfolio of two stocks. In most cases the neural networks performed on par with benchmarks, although some modelsExpand
Handling risk-on/risk-off dynamics with correlation regimes and correlation networks
In this paper, we present a framework for detecting distinct correlation regimes and analyzing the emerging state dependences for a multi-asset futures portfolio from 1998 to 2013. These correlationExpand