Paper: arXiv 2212.01807

Authors: Damian Kisiel, Denise Gorse

Abstract

Previous attempts to predict stock price from limit order book (LOB) data are mostly based on deep convolutional neural networks. Although convolutions offer efficiency by restricting their operations to local interactions, it is at the cost of potentially missing out on the detection of long-range dependencies. Recent studies address this problem by employing additional recurrent or attention layers that increase computational complexity. In this work, we propose Axial-LOB, a novel fully-attentional deep learning architecture for predicting price movements of stocks from LOB data. By utilizing gated position-sensitive axial attention layers our architecture is able to construct feature maps that incorporate global interactions, while significantly reducing the size of the parameter space. Unlike previous works, Axial-LOB does not rely on hand-crafted convolutional kernels and hence has stable performance under input permutations and the capacity to incorporate additional LOB features. The effectiveness of Axial-LOB is demonstrated on a large benchmark dataset, containing time series representations of millions of high-frequency trading events, where our model establishes a new state of the art, achieving an excellent directional classification performance at all tested prediction horizons.

Complexity vs Empirical Score

  • Math Complexity: 6.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper adapts axial attention mechanisms to Limit Order Book (LOB) data, striking a strong balance between mathematical deep learning formulation and high-frequency empirical validation. Demonstrating state-of-the-art results on a massive benchmark dataset underscores its empirical rigor and practical utility.

Research Flowchart

  flowchart TD
    A[Research Goal: Improve High-Frequency Price Prediction from LOB Data] --> B[Input Data: High-Frequency Limit Order Book LOB Time Series]
    B --> C[Methodology: Axial-LOB Architecture]
    C --> D[Gated Position-Sensitive Axial Attention Layers]
    D --> E[Construct Global Feature Maps without Hand-Crafted Conv Kernels]
    E --> F[Evaluation: Directional Classification Across Various Horizons]
    F --> G[Outcome: New SOTA Performance & Reduced Parameter Complexity]