Axial-LOB: High-Frequency Trading with Axial Attention

TL;DR

Axial-LOB introduces a fully-attentional neural network architecture utilizing axial attention for high-frequency stock price prediction from limit order book data, capturing global dependencies efficiently.

Contribution

It proposes a novel axial attention-based model that reduces complexity and improves performance over convolutional and recurrent methods in high-frequency trading prediction.

Findings

Achieves state-of-the-art accuracy on large benchmark dataset.

Demonstrates stable performance under input permutations.

Effectively incorporates additional LOB features.

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.