{"title":"深度限价订单预测:微观结构指南。","authors":"Antonio Briola, Silvia Bartolucci, Tomaso Aste","doi":"10.1080/14697688.2025.2522911","DOIUrl":null,"url":null,"abstract":"<p><p>We exploit cutting-edge deep learning methodologies to explore the predictability of high-frequency Limit Order Book mid-price changes for a heterogeneous set of stocks traded on the NASDAQ exchange. In so doing, we release 'LOBFrame', an open-source code base to efficiently process large-scale Limit Order Book data and quantitatively assess state-of-the-art deep learning models' forecasting capabilities. Our results are twofold. We demonstrate that the stocks' microstructural characteristics influence the efficacy of deep learning methods and that their high forecasting power does not necessarily correspond to actionable trading signals. We argue that traditional machine learning metrics fail to adequately assess the quality of forecasts in the Limit Order Book context. As an alternative, we propose an innovative operational framework that evaluates predictions' practicality by focusing on the probability of accurately forecasting complete transactions. This work offers academics and practitioners an avenue to make informed and robust decisions on the application of deep learning techniques, their scope and limitations, effectively exploiting emergent statistical properties of the Limit Order Book.</p>","PeriodicalId":20747,"journal":{"name":"Quantitative Finance","volume":" ","pages":"1-31"},"PeriodicalIF":1.4000,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12315853/pdf/","citationCount":"0","resultStr":"{\"title\":\"Deep limit order book forecasting: a microstructural guide.\",\"authors\":\"Antonio Briola, Silvia Bartolucci, Tomaso Aste\",\"doi\":\"10.1080/14697688.2025.2522911\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>We exploit cutting-edge deep learning methodologies to explore the predictability of high-frequency Limit Order Book mid-price changes for a heterogeneous set of stocks traded on the NASDAQ exchange. In so doing, we release 'LOBFrame', an open-source code base to efficiently process large-scale Limit Order Book data and quantitatively assess state-of-the-art deep learning models' forecasting capabilities. Our results are twofold. We demonstrate that the stocks' microstructural characteristics influence the efficacy of deep learning methods and that their high forecasting power does not necessarily correspond to actionable trading signals. We argue that traditional machine learning metrics fail to adequately assess the quality of forecasts in the Limit Order Book context. As an alternative, we propose an innovative operational framework that evaluates predictions' practicality by focusing on the probability of accurately forecasting complete transactions. This work offers academics and practitioners an avenue to make informed and robust decisions on the application of deep learning techniques, their scope and limitations, effectively exploiting emergent statistical properties of the Limit Order Book.</p>\",\"PeriodicalId\":20747,\"journal\":{\"name\":\"Quantitative Finance\",\"volume\":\" \",\"pages\":\"1-31\"},\"PeriodicalIF\":1.4000,\"publicationDate\":\"2025-07-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12315853/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Quantitative Finance\",\"FirstCategoryId\":\"96\",\"ListUrlMain\":\"https://doi.org/10.1080/14697688.2025.2522911\",\"RegionNum\":4,\"RegionCategory\":\"经济学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"BUSINESS, FINANCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Quantitative Finance","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.1080/14697688.2025.2522911","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}
Deep limit order book forecasting: a microstructural guide.
We exploit cutting-edge deep learning methodologies to explore the predictability of high-frequency Limit Order Book mid-price changes for a heterogeneous set of stocks traded on the NASDAQ exchange. In so doing, we release 'LOBFrame', an open-source code base to efficiently process large-scale Limit Order Book data and quantitatively assess state-of-the-art deep learning models' forecasting capabilities. Our results are twofold. We demonstrate that the stocks' microstructural characteristics influence the efficacy of deep learning methods and that their high forecasting power does not necessarily correspond to actionable trading signals. We argue that traditional machine learning metrics fail to adequately assess the quality of forecasts in the Limit Order Book context. As an alternative, we propose an innovative operational framework that evaluates predictions' practicality by focusing on the probability of accurately forecasting complete transactions. This work offers academics and practitioners an avenue to make informed and robust decisions on the application of deep learning techniques, their scope and limitations, effectively exploiting emergent statistical properties of the Limit Order Book.
期刊介绍:
The frontiers of finance are shifting rapidly, driven in part by the increasing use of quantitative methods in the field. Quantitative Finance welcomes original research articles that reflect the dynamism of this area. The journal provides an interdisciplinary forum for presenting both theoretical and empirical approaches and offers rapid publication of original new work with high standards of quality. The readership is broad, embracing researchers and practitioners across a range of specialisms and within a variety of organizations. All articles should aim to be of interest to this broad readership.