HLOB -- Information Persistence and Structure in Limit Order Books

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Briola, Antonio, Bartolucci, Silvia, Aste, Tomaso
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916273719345152
author Briola, Antonio
Bartolucci, Silvia
Aste, Tomaso
author_facet Briola, Antonio
Bartolucci, Silvia
Aste, Tomaso
contents We introduce a novel large-scale deep learning model for Limit Order Book mid-price changes forecasting, and we name it `HLOB'. This architecture (i) exploits the information encoded by an Information Filtering Network, namely the Triangulated Maximally Filtered Graph, to unveil deeper and non-trivial dependency structures among volume levels; and (ii) guarantees deterministic design choices to handle the complexity of the underlying system by drawing inspiration from the groundbreaking class of Homological Convolutional Neural Networks. We test our model against 9 state-of-the-art deep learning alternatives on 3 real-world Limit Order Book datasets, each including 15 stocks traded on the NASDAQ exchange, and we systematically characterize the scenarios where HLOB outperforms state-of-the-art architectures. Our approach sheds new light on the spatial distribution of information in Limit Order Books and on its degradation over increasing prediction horizons, narrowing the gap between microstructural modeling and deep learning-based forecasting in high-frequency financial markets.
format Preprint
id arxiv_https___arxiv_org_abs_2405_18938
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle HLOB -- Information Persistence and Structure in Limit Order Books
Briola, Antonio
Bartolucci, Silvia
Aste, Tomaso
Trading and Market Microstructure
Machine Learning
We introduce a novel large-scale deep learning model for Limit Order Book mid-price changes forecasting, and we name it `HLOB'. This architecture (i) exploits the information encoded by an Information Filtering Network, namely the Triangulated Maximally Filtered Graph, to unveil deeper and non-trivial dependency structures among volume levels; and (ii) guarantees deterministic design choices to handle the complexity of the underlying system by drawing inspiration from the groundbreaking class of Homological Convolutional Neural Networks. We test our model against 9 state-of-the-art deep learning alternatives on 3 real-world Limit Order Book datasets, each including 15 stocks traded on the NASDAQ exchange, and we systematically characterize the scenarios where HLOB outperforms state-of-the-art architectures. Our approach sheds new light on the spatial distribution of information in Limit Order Books and on its degradation over increasing prediction horizons, narrowing the gap between microstructural modeling and deep learning-based forecasting in high-frequency financial markets.
title HLOB -- Information Persistence and Structure in Limit Order Books
topic Trading and Market Microstructure
Machine Learning
url https://arxiv.org/abs/2405.18938