Predicting Price Movements in High-Frequency Financial Data with Spiking Neural Networks

2025-12-05

AI Summary

Traditional models struggle to effectively capture millisecond-scale price spikes in high-frequency trading (HFT). Method: This paper proposes a novel spiking neural network (SNN) framework that encodes stock price time series into temporally precise spike trains, incorporates an inhibitory competitive mechanism to enhance temporal discriminability, and introduces a penalty-aware spike accuracy (PSA) metric as a task-specific optimization objective. The model combines unsupervised spike-timing-dependent plasticity (STDP) pretraining with supervised backpropagation-based fine-tuning, and employs Bayesian hyperparameter optimization for robustness. Contribution/Results: Evaluated on real-world HFT data, the method achieves a cumulative return of 76.8%, substantially outperforming benchmark models (42.54%). This work provides the first empirical validation of biologically inspired, dynamically adaptive SNNs for both predicting rapid price discontinuities and executing profitable live trading strategies in HFT environments.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Learning to Search
• Machine Learning: Hardware-aware ML

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Social networks and social learning
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Modern high-frequency trading (HFT) environments are characterized by sudden price spikes that present both risk and opportunity, but conventional financial models often fail to capture the required fine temporal structure. Spiking Neural Networks (SNNs) offer a biologically inspired framework well-suited to these challenges due to their natural ability to process discrete events and preserve millisecond-scale timing. This work investigates the application of SNNs to high-frequency price-spike forecasting, enhancing performance via robust hyperparameter tuning with Bayesian Optimization (BO). This work converts high-frequency stock data into spike trains and evaluates three architectures: an established unsupervised STDP-trained SNN, a novel SNN with explicit inhibitory competition, and a supervised backpropagation network. BO was driven by a novel objective, Penalized Spike Accuracy (PSA), designed to ensure a network's predicted price spike rate aligns with the empirical rate of price events. Simulated trading demonstrated that models optimized with PSA consistently outperformed their Spike Accuracy (SA)-tuned counterparts and baselines. Specifically, the extended SNN model with PSA achieved the highest cumulative return (76.8%) in simple backtesting, significantly surpassing the supervised alternative (42.54% return). These results validate the potential of spiking networks, when robustly tuned with task-specific objectives, for effective price spike forecasting in HFT.

Problem

• Predict price movements in high-frequency trading data
• Apply spiking neural networks to forecast price spikes
• Enhance forecasting with Bayesian optimization and novel objectives

Innovation

• Spiking Neural Networks process high-frequency financial data
• Bayesian Optimization tunes hyperparameters with Penalized Spike Accuracy
• Novel SNN architecture with inhibitory competition outperforms baselines