Temporal Attention augmented Bilinear Network for Financial Time-Series Data Analysis

TL;DR

This paper introduces a neural network layer combining bilinear projection and attention mechanisms, improving the accuracy and interpretability of financial time-series forecasting, especially for high-frequency trading applications.

Contribution

It proposes a novel neural network layer that enhances temporal focus and interpretability in financial data analysis, outperforming deeper models with fewer computations.

Findings

Outperforms state-of-the-art models on large-scale LOB data

Requires fewer computations than deeper architectures

Provides interpretability by highlighting important time instances

Abstract

Financial time-series forecasting has long been a challenging problem because of the inherently noisy and stochastic nature of the market. In the High-Frequency Trading (HFT), forecasting for trading purposes is even a more challenging task since an automated inference system is required to be both accurate and fast. In this paper, we propose a neural network layer architecture that incorporates the idea of bilinear projection as well as an attention mechanism that enables the layer to detect and focus on crucial temporal information. The resulting network is highly interpretable, given its ability to highlight the importance and contribution of each temporal instance, thus allowing further analysis on the time instances of interest. Our experiments in a large-scale Limit Order Book (LOB) dataset show that a two-hidden-layer network utilizing our proposed layer outperforms by a large margin all existing state-of-the-art results coming from much deeper architectures while requiring far fewer computations.