AI-Powered Energy Algorithmic Trading: Integrating Hidden Markov Models with Neural Networks

TL;DR

This study proposes a dual-model alpha-generation framework that fuses Hidden Markov Models (HMM) with neural networks to predict price movements, integrated with Black-Litterman portfolio optimization. Implemented on QuantConnect with a three-year warm-up, the system uses a consensus rule requiring aligned signals before trading, focusing on liquid energy stocks. In backtests over 2019–2022 (including the COVID period), the approach achieved about $83\%$ return with a Sharpe ratio of $0.77$, while incorporating two risk models for drawdown control. The work demonstrates that coupling regime-aware HMMs with nonlinear pattern learning improves signal reliability and risk-adjusted performance, and it underscores the importance of reproducibility via the QuantConnect LEAN framework. Potential extensions include wavelet preprocessing, Monte Carlo generalization tests, hyperparameter tuning, and broader sector diversification.

Abstract

In quantitative finance, machine learning methods are essential for alpha generation. This study introduces a new approach that combines Hidden Markov Models (HMM) and neural networks, integrated with Black-Litterman portfolio optimization. During the COVID period (2019-2022), this dual-model approach achieved a 83% return with a Sharpe ratio of 0.77. It incorporates two risk models to enhance risk management, showing efficiency during volatile periods. The methodology was implemented on the QuantConnect platform, which was chosen for its robust framework and experimental reproducibility. The system, which predicts future price movements, includes a three-year warm-up to ensure proper algorithm function. It targets highly liquid, large-cap energy stocks to ensure stable and predictable performance while also considering broker payments. The dual-model alpha system utilizes log returns to select the optimal state based on the historical performance. It combines state predictions with neural network outputs, which are based on historical data, to generate trading signals. This study examined the architecture of the trading system, data pre-processing, training, and performance. The full code and backtesting data are available under the QuantConnect terms: https://github.com/tiagomonteiro0715/AI-Powered-Energy-Algorithmic-Trading-Integrating-Hidden-Markov-Models-with-Neural-Networks

Figures (4)

• Figure 1: Workflow of Algorithmic Trading System
• Figure 2: HMM Architecture for Simulating Stock Price Fluctuations
• Figure 3: Neural Network Architecture for Detecting Complex Patterns in Historical Price Data
• Figure 4: Cumulative Returns Comparison (January 2019--January 2022)