eprintid: 10204213 rev_number: 6 eprint_status: archive userid: 699 dir: disk0/10/20/42/13 datestamp: 2025-02-03 11:50:23 lastmod: 2025-02-03 11:50:23 status_changed: 2025-02-03 11:50:23 type: article metadata_visibility: show sword_depositor: 699 creators_name: Briola, A creators_name: Bartolucci, S creators_name: Aste, T title: HLOB–Information persistence and structure in limit order books ispublished: pub divisions: UCL divisions: B04 divisions: F48 note: © 2024 The Authors. Published by Elsevier Ltd. under a Creative Commons license (http://creativecommons.org/licenses/by/4.0/). abstract: 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. date: 2025-03-25 date_type: published publisher: Elsevier BV official_url: https://doi.org/10.1016/j.eswa.2024.126078 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2343322 doi: 10.1016/j.eswa.2024.126078 lyricists_name: Aste, Tomaso lyricists_name: Bartolucci, Silvia lyricists_id: TASTE72 lyricists_id: SBART13 actors_name: Bartolucci, Silvia actors_id: SBART13 actors_role: owner full_text_status: public publication: Expert Systems with Applications volume: 266 article_number: 126078 citation: Briola, A; Bartolucci, S; Aste, T; (2025) HLOB–Information persistence and structure in limit order books. Expert Systems with Applications , 266 , Article 126078. 10.1016/j.eswa.2024.126078 <https://doi.org/10.1016/j.eswa.2024.126078>. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10204213/1/1-s2.0-S0957417424029452-main.pdf