{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T02:02:50Z","timestamp":1793066570635,"version":"4.5.0"},"reference-count":100,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2023,7,21]],"date-time":"2023-07-21T00:00:00Z","timestamp":1689897600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Mathematical Finance"],"published-print":{"date-parts":[[2023,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>We employ deep learning in forecasting high\u2010frequency returns at multiple horizons for 115 stocks traded on Nasdaq using order book information at the most granular level. While raw order book states can be used as input to the forecasting models, we achieve state\u2010of\u2010the\u2010art predictive accuracy by training simpler \u201coff\u2010the\u2010shelf\u201d artificial neural networks on stationary inputs derived from the order book. Specifically, models trained on order flow significantly outperform most models trained directly on order books. Using cross\u2010sectional regressions, we link the forecasting performance of a long short\u2010term memory network to stock characteristics at the market microstructure level, suggesting that \u201cinformation\u2010rich\u201d stocks can be predicted more accurately. Finally, we demonstrate that the effective horizon of stock specific forecasts is approximately two average price\u00a0changes.<\/jats:p>","DOI":"10.1111\/mafi.12413","type":"journal-article","created":{"date-parts":[[2023,7,21]],"date-time":"2023-07-21T07:51:44Z","timestamp":1689925904000},"page":"1044-1081","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":39,"title":["Deep order flow imbalance: Extracting alpha at multiple horizons from the limit order book"],"prefix":"10.1111","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6794-4335","authenticated-orcid":false,"given":"Petter N.","family":"Kolm","sequence":"first","affiliation":[{"name":"Courant Institute of Mathematical Sciences New York University New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeremy","family":"Turiel","sequence":"additional","affiliation":[{"name":"Department of Computer Science University College London London UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicholas","family":"Westray","sequence":"additional","affiliation":[{"name":"Courant Institute of Mathematical Sciences New York University New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2023,7,21]]},"reference":[{"key":"e_1_2_8_2_1","volume-title":"TensorFlow: Large\u2010scale machine learning on heterogeneous systems","author":"Abadi M.","year":"2015"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9781316683040"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(80)90020-3"},{"key":"e_1_2_8_5_1","unstructured":"Arjovsky M. Bottou L. Gulrajani I. &Lopez\u2010Paz D.(2020).Invariant risk minimization.arXiv preprint arXiv:1907.02893."},{"key":"e_1_2_8_6_1","unstructured":"Bahdanau D. Cho K. &Bengio Y.(2014).Neural machine translation by jointly learning to align and translate.arXiv preprint arXiv:1409.0473."},{"key":"e_1_2_8_7_1","unstructured":"Bengio Y.(2012).Deep learning of representations for unsupervised and transfer learning. InProceedings of ICML Workshop on Unsupervised and Transfer Learning JMLR Workshop and Conference Proceedings(pp. 17\u201336)."},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1017\/9781316659335"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697680400000022"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697680500397623"},{"key":"e_1_2_8_11_1","doi-asserted-by":"crossref","unstructured":"Briola A. Turiel J. &Aste T.(2020).Deep learning modeling of limit order book: A comparative perspective.arXiv preprint arXiv:2007.07319.","DOI":"10.2139\/ssrn.3714230"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1111\/jofi.12769"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"Bugaenko A.(2020).Empirical study of market impact conditional on order\u2010flow imbalance.","DOI":"10.2139\/ssrn.3589220"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/fut.20334"},{"key":"e_1_2_8_15_1","doi-asserted-by":"crossref","unstructured":"Cho K. Van Merri\u00ebnboer B. Gulcehre C. Bahdanau D. Bougares F. Schwenk H. &Bengio Y.(2014).Learning phrase representations using RNN encoder\u2010decoder for statistical machine translation.arXiv preprint arXiv:1406.1078.","DOI":"10.3115\/v1\/D14-1179"},{"key":"e_1_2_8_16_1","volume-title":"Keras","author":"Chollet F.","year":"2015"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0304-405X(02)00136-8"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0304-405X(03)00175-2"},{"key":"e_1_2_8_19_1","unstructured":"Clevert D.\u2010A. Unterthiner T. &Hochreiter S.(2015).Fast and accurate deep network learning by exponential linear units (ELUs).arXiv preprint arXiv:1511.07289."},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1080\/713665670"},{"key":"e_1_2_8_21_1","doi-asserted-by":"crossref","unstructured":"Cont R. Cucuringu M. &Zhang C.(2021).Price impact of order flow imbalance: Multi\u2010level cross\u2010sectional and forecasting.Quantitative Finance.arXiv preprint arXiv:2112.13213 to appear.","DOI":"10.2139\/ssrn.3993561"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1093\/jjfinec\/nbt003"},{"key":"e_1_2_8_23_1","doi-asserted-by":"crossref","unstructured":"Curato G. &Lillo F.(2015).How tick size affects the high frequency scaling of stock return distributions. InFinancial econometrics and empirical market microstructure(pp. 55\u201376). Springer.","DOI":"10.1007\/978-3-319-09946-0_6"},{"key":"e_1_2_8_24_1","unstructured":"Devlin J. Chang M.\u2010W. Lee K. &Toutanova K.(2018).BERT: Pre\u2010training of deep bidirectional transformers for language understanding.arXiv preprint arXiv:1810.04805."},{"key":"e_1_2_8_25_1","doi-asserted-by":"crossref","unstructured":"Donahue J. Hendricks L. A. Guadarrama S. Rohrbach M. Venugopalan S. Saenko K. &Darrell T.(2015).Long\u2010term recurrent convolutional networks for visual recognition and description. InProceedings of the IEEE Conference on Computer Vision And Pattern Recognition(pp. 2625\u20132634).","DOI":"10.1109\/CVPR.2015.7298878"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(87)90029-8"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2010.528444"},{"key":"e_1_2_8_28_1","doi-asserted-by":"crossref","unstructured":"Elmsili B. &Outtaj B.(2018).Artificial neural networks applications in economics and management research: An exploratory literature review. In2018 4th International Conference on Optimization and Applications (ICOA)(pp. 1\u20136). IEEE.","DOI":"10.1109\/ICOA.2018.8370600"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10618-019-00619-1"},{"key":"e_1_2_8_30_1","volume-title":"The elements of statistical learning. Series in Statistics New York","author":"Friedman J.","year":"2017"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(76)90006-4"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1162\/089976600300015015"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(85)90044-3"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfineco.2008.08.002"},{"key":"e_1_2_8_35_1","volume-title":"Deep learning","author":"Goodfellow I.","year":"2016"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1142\/S2382626616500064"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2013.803148"},{"key":"e_1_2_8_38_1","unstructured":"Graves A.(2013).Generating sequences with recurrent neural networks.arXiv preprint arXiv:1308.0850."},{"key":"e_1_2_8_39_1","doi-asserted-by":"crossref","unstructured":"Graves A. Mohamed A.\u2010R. &Hinton G.(2013).Speech recognition with deep recurrent neural networks. In2013 IEEE International Conference on Acoustics Speech and Signal Processing(pp. 6645\u20136649). IEEE.","DOI":"10.1109\/ICASSP.2013.6638947"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2016.2582924"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1515\/9780691218632"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(88)90070-0"},{"key":"e_1_2_8_43_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1540-6261.1991.tb03749.x"},{"key":"e_1_2_8_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.finmar.2013.05.003"},{"key":"e_1_2_8_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(81)90020-9"},{"key":"e_1_2_8_46_1","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"e_1_2_8_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfineco.2014.04.002"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(91)90009-T"},{"key":"e_1_2_8_49_1","doi-asserted-by":"crossref","unstructured":"Huang R. &Polak T.(2011).LOBSTER: Limit order book reconstruction system.SSRN 1977207.","DOI":"10.2139\/ssrn.1977207"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1093\/rfs\/7.1.179"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1214\/15-AAP1164"},{"key":"e_1_2_8_52_1","unstructured":"Jozefowicz R. Zaremba W. &Sutskever I.(2015).An empirical exploration of recurrent network architectures. InInternational Conference on Machine Learning(pp. 2342\u20132350). PMLR."},{"key":"e_1_2_8_53_1","doi-asserted-by":"publisher","DOI":"10.1086\/503651"},{"key":"e_1_2_8_54_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2015.1032546"},{"key":"e_1_2_8_55_1","unstructured":"Keskar N. S. Mudigere D. Nocedal J. Smelyanskiy M. &Tang P. T. P.(2017).on large\u2010batch training for deep learning: Generalization gap and sharp minima."},{"key":"e_1_2_8_56_1","doi-asserted-by":"crossref","unstructured":"Kim Y. Jernite Y. Sontag D. &Rush A.(2016).Character\u2010aware neural language models. InProceedings of the AAAI Conference on Artificial Intelligence(Vol. 30).","DOI":"10.1609\/aaai.v30i1.10362"},{"key":"e_1_2_8_57_1","unstructured":"Kingma D. P. &Ba J.(2014).Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980."},{"key":"e_1_2_8_58_1","unstructured":"Kolm P. N. &Westray N.(2023).Information content of cross\u2010sectional and multilevel order flow imbalances: A Bayesian approach. In preparation."},{"key":"e_1_2_8_59_1","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky A.","year":"2012","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_8_60_1","doi-asserted-by":"publisher","DOI":"10.2307\/1913210"},{"key":"e_1_2_8_61_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_8_62_1","doi-asserted-by":"crossref","unstructured":"Li Y. &Ma W.(2010).Applications of artificial neural networks in financial economics: A survey. In2010 International Symposium on Computational Intelligence and Design(Vol. 1) pp. 211\u2013214. IEEE.","DOI":"10.1109\/ISCID.2010.70"},{"key":"e_1_2_8_63_1","doi-asserted-by":"publisher","DOI":"10.2202\/1558-3708.1226"},{"key":"e_1_2_8_64_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2015.02.008"},{"key":"e_1_2_8_65_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2959655"},{"key":"e_1_2_8_66_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2019.1634277"},{"key":"e_1_2_8_67_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2021.1947511"},{"key":"e_1_2_8_68_1","unstructured":"Mnih V. Heess N. Graves A. &Kavukcuoglu K.(2014).Recurrent models of visual attention.arXiv preprint arXiv:1406.6247."},{"key":"e_1_2_8_69_1","doi-asserted-by":"publisher","DOI":"10.1002\/for.2543"},{"key":"e_1_2_8_70_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2020.106384"},{"key":"e_1_2_8_71_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2009.191"},{"key":"e_1_2_8_72_1","unstructured":"Pascanu R. Gulcehre C. Cho K. &Bengio Y.(2013).How to construct deep recurrent neural networks.arXiv preprint arXiv:1312.6026."},{"key":"e_1_2_8_73_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2020.06.006"},{"key":"e_1_2_8_74_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2010.12.010"},{"key":"e_1_2_8_75_1","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511803161"},{"key":"e_1_2_8_76_1","unstructured":"Rahimikia E. &Poon S.\u2010H.(2023).Machine learning for realised volatility forecasting.SSRN 3707796."},{"key":"e_1_2_8_77_1","doi-asserted-by":"publisher","DOI":"10.1038\/323533a0"},{"key":"e_1_2_8_78_1","doi-asserted-by":"crossref","unstructured":"Sainath T. N. Vinyals O. Senior A. &Sak H.(2015).Convolutional long short\u2010term memory fully connected deep neural networks. In2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)(pp. 4580\u20134584). IEEE.","DOI":"10.1109\/ICASSP.2015.7178838"},{"key":"e_1_2_8_79_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2014.09.003"},{"key":"e_1_2_8_80_1","doi-asserted-by":"crossref","unstructured":"Sermanet P. &LeCun Y.(2011).Traffic sign recognition with multi\u2010scale convolutional networks. InProceedings of the 2011 International Joint Conference on Neural Networks(pp. 2809\u20132813). IEEE.","DOI":"10.1109\/IJCNN.2011.6033589"},{"key":"e_1_2_8_81_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2020.106181"},{"key":"e_1_2_8_82_1","unstructured":"Shi X. Chen Z. Wang H. Yeung D.\u2010Y. Wong W.\u2010K. &Woo W.\u2010C.(2015).Convolutional LSTM network: A machine learning approach for precipitation nowcasting. InProceedings of the Advances in Neural Information Processing Systems 28."},{"key":"e_1_2_8_83_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2018.1546053"},{"key":"e_1_2_8_84_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2019.1622295"},{"key":"e_1_2_8_85_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1540-6261.1978.tb02053.x"},{"key":"e_1_2_8_86_1","volume-title":"The bitter lesson","author":"Sutton R.","year":"2019"},{"key":"e_1_2_8_87_1","doi-asserted-by":"crossref","unstructured":"Szegedy C. Liu W. Jia Y. Sermanet P. Reed S. Anguelov D. Erhan D. Vanhoucke V. &Rabinovich A.(2015).Going deeper with convolutions. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition(pp. 1\u20139).","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_8_88_1","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2012.690886"},{"key":"e_1_2_8_89_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2018.2869225"},{"key":"e_1_2_8_90_1","doi-asserted-by":"crossref","unstructured":"Tsantekidis A. Passalis N. Tefas A. Kanniainen J. Gabbouj M. &Iosifidis A.(2017).Forecasting stock prices from the limit order book using convolutional neural networks. In2017 IEEE 19th Conference on Business Informatics (CBI)(Vol. 1 pp. 7\u201312). IEEE.","DOI":"10.1109\/CBI.2017.23"},{"key":"e_1_2_8_91_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2020.106401"},{"key":"e_1_2_8_92_1","doi-asserted-by":"crossref","unstructured":"Vinyals O. Toshev A. Bengio S. &Erhan D.(2015).Show and tell: A neural image caption generator. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition(pp. 3156\u20133164).","DOI":"10.1109\/CVPR.2015.7298935"},{"key":"e_1_2_8_93_1","doi-asserted-by":"publisher","DOI":"10.2307\/1912934"},{"key":"e_1_2_8_94_1","unstructured":"Wilson A. G. &Izmailov P.(2020).Bayesian deep learning and a probabilistic perspective of generalization.arXiv preprint arXiv:2002.08791."},{"key":"e_1_2_8_95_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0378-7206(98)00050-0"},{"key":"e_1_2_8_96_1","doi-asserted-by":"publisher","DOI":"10.1142\/S2382626619500114"},{"key":"e_1_2_8_97_1","doi-asserted-by":"publisher","DOI":"10.1080\/1350486X.2021.1967767"},{"key":"e_1_2_8_98_1","unstructured":"Zhang Z. &Zohren S.(2021).Multi\u2010horizon forecasting for limit order books: Novel deep learning approaches and hardware acceleration using intelligent processing units.arXiv preprint arXiv:2105.10430."},{"key":"e_1_2_8_99_1","unstructured":"Zhang Z. Zohren S. &Roberts S.(2018).BDLOB: Bayesian deep convolutional neural networks for limit order books.arXiv preprint arXiv:1811.10041."},{"key":"e_1_2_8_100_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2019.2907260"},{"key":"e_1_2_8_101_1","unstructured":"Zhang Z. Zohren S. &Roberts S.(2019b).Extending deep learning models for limit order books to quantile regression.arXiv preprint arXiv:1906.04404."}],"container-title":["Mathematical Finance"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/mafi.12413","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,9,13]],"date-time":"2023-09-13T03:07:04Z","timestamp":1694574424000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/mafi.12413"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,7,21]]},"references-count":100,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2023,10]]}},"alternative-id":["10.1111\/mafi.12413"],"URL":"https:\/\/doi.org\/10.1111\/mafi.12413","archive":["Portico"],"relation":{},"ISSN":["0960-1627","1467-9965"],"issn-type":[{"value":"0960-1627","type":"print"},{"value":"1467-9965","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,7,21]]},"assertion":[{"value":"2021-12-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-06-10","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2023-07-21","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}