Skip to content
All library documents

Deep Q-Learning for Multi-Asset Portfolio Trading

Article arXiv papers · Author: Hyungjun Park et al.
Read the original

Summary

This study formulates portfolio trading as a Markov decision process and trains an agent with deep Q-learning to choose allocations across multiple assets. Its action space is discrete and combinatorial: for each asset, the agent selects a trading direction at a predefined size. To address actions that violate constraints, the method maps an infeasible proposal to the closest feasible alternative. It also describes an agent and Q-network design intended to manage the multi-asset action space and simulate feasible actions in each state.

The approach is evaluated through backtests on two representative portfolios, with results reported as superior to benchmark strategies. The document does not identify the portfolios, benchmarks, evaluation period, or transaction-cost assumptions, so the breadth and live-trading relevance of that comparison cannot be judged from the description alone.

Key ideas

  • The portfolio decision process is modeled as a Markov decision process trained with deep Q-learning.
  • The agent selects discrete directions and predefined trading sizes for each asset.
  • A mapping step converts infeasible proposed actions into nearby feasible actions.
  • Backtests on two portfolios are reported to outperform benchmark strategies, but evaluation details are not supplied.

Tags

Machine learningPortfolio constructionBacktestingPosition sizing
Full text
# An intelligent financial portfolio trading strategy using deep Q-learning


# An intelligent financial portfolio trading strategy using deep Q-learning









Portfolio traders strive to identify dynamic portfolio allocation schemes so that their total budgets are efficiently allocated through the investment horizon. This study proposes a novel portfolio trading strategy in which an intelligent agent is trained to identify an optimal trading action by using deep Q-learning. We formulate a Markov decision process model for the portfolio trading process, and the model adopts a discrete combinatorial action space, determining the trading direction at prespecified trading size for each asset, to ensure practical applicability. Our novel portfolio trading strategy takes advantage of three features to outperform in real-world trading. First, a mapping function is devised to handle and transform an initially found but infeasible action into a feasible action closest to the originally proposed ideal action. Second, by overcoming the dimensionality problem, this study establishes models of agent and Q-network for deriving a multi-asset trading strategy in the predefined action space. Last, this study introduces a technique that has the advantage of deriving a well-fitted multi-asset trading strategy by designing an agent to simulate all feasible actions in each state. To validate our approach, we conduct backtests for two representative portfolios and demonstrate superior results over the benchmark strategies.

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.