Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep Reinforcement Learning for Investor-Specific Portfolio Optimization: A Volatility-Guided Asset Selection Approach

Arishi Orra, Aryan Bhambu, Himanshu Choudhary, Manoj Thakur, Selvaraju Natarajan

TL;DR

The paper addresses investor-specific portfolio optimization under dynamic markets by combining volatility-guided asset pre-selection with deep reinforcement learning. It uses the volatility forecast from a $GARCH(1,1)$ model to classify Dow Jones 30 stocks into Aggressive, Moderate, and Conservative pools, then trains a Proximal Policy Optimization (PPO) agent to learn allocation policies within these pools. On historical data, the approach yields portfolios that substantially outperform traditional baselines in risk-adjusted terms, with the Moderate-DRL portfolio offering the best balance of return and risk. The work highlights the practical value of aligning asset choice with investor risk preferences and suggests directions for extending to multiple asset classes and incorporating richer risk measures like CVaR.

Abstract

Portfolio optimization requires dynamic allocation of funds by balancing the risk and return tradeoff under dynamic market conditions. With the recent advancements in AI, Deep Reinforcement Learning (DRL) has gained prominence in providing adaptive and scalable strategies for portfolio optimization. However, the success of these strategies depends not only on their ability to adapt to market dynamics but also on the careful pre-selection of assets that influence overall portfolio performance. Incorporating the investor's preference in pre-selecting assets for a portfolio is essential in refining their investment strategies. This study proposes a volatility-guided DRL-based portfolio optimization framework that dynamically constructs portfolios based on investors' risk profiles. The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model is utilized for volatility forecasting of stocks and categorizes them based on their volatility as aggressive, moderate, and conservative. The DRL agent is then employed to learn an optimal investment policy by interacting with the historical market data. The efficacy of the proposed methodology is established using stocks from the Dow $30$ index. The proposed investor-specific DRL-based portfolios outperformed the baseline strategies by generating consistent risk-adjusted returns.

Deep Reinforcement Learning for Investor-Specific Portfolio Optimization: A Volatility-Guided Asset Selection Approach

TL;DR

The paper addresses investor-specific portfolio optimization under dynamic markets by combining volatility-guided asset pre-selection with deep reinforcement learning. It uses the volatility forecast from a model to classify Dow Jones 30 stocks into Aggressive, Moderate, and Conservative pools, then trains a Proximal Policy Optimization (PPO) agent to learn allocation policies within these pools. On historical data, the approach yields portfolios that substantially outperform traditional baselines in risk-adjusted terms, with the Moderate-DRL portfolio offering the best balance of return and risk. The work highlights the practical value of aligning asset choice with investor risk preferences and suggests directions for extending to multiple asset classes and incorporating richer risk measures like CVaR.

Abstract

Portfolio optimization requires dynamic allocation of funds by balancing the risk and return tradeoff under dynamic market conditions. With the recent advancements in AI, Deep Reinforcement Learning (DRL) has gained prominence in providing adaptive and scalable strategies for portfolio optimization. However, the success of these strategies depends not only on their ability to adapt to market dynamics but also on the careful pre-selection of assets that influence overall portfolio performance. Incorporating the investor's preference in pre-selecting assets for a portfolio is essential in refining their investment strategies. This study proposes a volatility-guided DRL-based portfolio optimization framework that dynamically constructs portfolios based on investors' risk profiles. The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model is utilized for volatility forecasting of stocks and categorizes them based on their volatility as aggressive, moderate, and conservative. The DRL agent is then employed to learn an optimal investment policy by interacting with the historical market data. The efficacy of the proposed methodology is established using stocks from the Dow index. The proposed investor-specific DRL-based portfolios outperformed the baseline strategies by generating consistent risk-adjusted returns.
Paper Structure (7 sections, 2 equations, 2 figures, 1 table)

This paper contains 7 sections, 2 equations, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Proposed Methodology
  4. Experiments
  5. Data Description and Experimental settings
  6. Results and Discussion
  7. Conclusion

Figures (2)

  • Figure 1: The schematic diagram depicting the workflow of the proposed methodology.
  • Figure 2: Cumulative return plots of the proposed methodology against the benchmark portfolio strategies over the trading period.