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

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.

12 pages, 4 figures, 3 tables

References in corpus (10)

• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Recurrent Models of Visual Attention
• Learning to Diagnose with LSTM Recurrent Neural Networks
• A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction
• ABC-CNN: An Attention Based Convolutional Neural Network for Visual Question Answering
• Tensor Representation in High-Frequency Financial Data for Price Change Prediction
• Multilinear Class-Specific Discriminant Analysis
• Deep Portfolio Theory
• Learning Deep Matrix Representations
• Using machine learning for medium frequency derivative portfolio trading

Cited by in corpus (17)

• Applications of deep learning in stock market prediction: recent progress
• A General Survey on Attention Mechanisms in Deep Learning
• Attention in Natural Language Processing
• DeepLOB: Deep Convolutional Neural Networks for Limit Order Books
• Self-supervised Autoregressive Domain Adaptation for Time Series Data
• A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting
• Multilinear Compressive Learning
• Low-Rank Temporal Attention-Augmented Bilinear Network for financial time-series forecasting
• Transformers for Limit Order Books
• Machine Learning for Forecasting Mid Price Movement using Limit Order Book Data
• Open-domain Event Extraction and Embedding for Natural Gas Market Prediction
• Multi-Horizon Forecasting for Limit Order Books: Novel Deep Learning Approaches and Hardware Acceleration using Intelligent Processing Units
• The Arrival of News and Return Jumps in Stock Markets: A Nonparametric Approach
• Bilinear Input Normalization for Neural Networks in Financial Forecasting
• Knowledge Distillation By Sparse Representation Matching
• DeepTimeAnomalyViz: A Tool for Visualizing and Post-processing Deep Learning Anomaly Detection Results for Industrial Time-Series
• Visualising Deep Network's Time-Series Representations