{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T03:06:13Z","timestamp":1780455973828,"version":"3.54.1"},"reference-count":52,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2026,2,1]],"date-time":"2026-02-01T00:00:00Z","timestamp":1769904000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"},{"start":{"date-parts":[[2026,2,1]],"date-time":"2026-02-01T00:00:00Z","timestamp":1769904000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001871","name":"Funda\u00e7\u00e3o para a Ci\u00eancia e a Tecnologia","doi-asserted-by":"publisher","award":["LA\/P\/0063\/2020"],"award-info":[{"award-number":["LA\/P\/0063\/2020"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001871","name":"Funda\u00e7\u00e3o para a Ci\u00eancia e a Tecnologia","doi-asserted-by":"publisher","award":["UID\/4105\/2025"],"award-info":[{"award-number":["UID\/4105\/2025"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001871","name":"Funda\u00e7\u00e3o para a Ci\u00eancia e a Tecnologia","doi-asserted-by":"publisher","award":["2023.01070.BD"],"award-info":[{"award-number":["2023.01070.BD"]}],"id":[{"id":"10.13039\/501100001871","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Journal of Forecasting"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>Forecasting the covariance matrix of asset returns is central to portfolio construction, risk management, and asset pricing. However, most existing models struggle at medium\u2010term horizons, several weeks to months, where shifting market regimes and slower dynamics prevail. We propose a novel deep learning framework that integrates three\u2010dimensional convolutional neural networks, bidirectional long short\u2010term memory, and multihead attention to capture complex spatiotemporal patterns in asset return dynamics. Using daily data on 14 exchange\u2010traded funds from 2017 to 2023, we demonstrate that our model improves out\u2010of\u2010sample covariance forecasts by reducing Euclidean and Frobenius distance metrics by up to 20% compared with classical benchmarks such as shrinkage estimators and GARCH\u2010type models. These gains persist across distinct market regimes, including bull and bear periods, and remain robust across various forecast horizons and under both raw and excess return specifications. Portfolio simulations based on global minimum variance strategies reveal that the proposed model consistently delivers lower volatility and moderate turnover, even under no\u2010short\u2010selling constraints. This balance between risk reduction and trading efficiency underscores the economic relevance of the forecasts, particularly for institutional investors managing portfolios at medium\u2010term horizons.<\/jats:p>","DOI":"10.1002\/for.70110","type":"journal-article","created":{"date-parts":[[2026,2,2]],"date-time":"2026-02-02T01:01:49Z","timestamp":1769994109000},"page":"1797-1828","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Deep Learning Framework for Forecasting Medium\u2010Term Covariance in Multiasset Portfolios"],"prefix":"10.1002","volume":"45","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-2912-8364","authenticated-orcid":false,"given":"Pedro","family":"Reis","sequence":"first","affiliation":[{"name":"School of Economics and Management and Center for Economics and Finance (cef.UP) University of Porto  Porto Portugal"},{"name":"INESCTEC Campus da Faculdade de Engenharia da Universidade do Porto  Porto Portugal"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4312-0451","authenticated-orcid":false,"given":"Ana","family":"Paula\u00a0Serra","sequence":"additional","affiliation":[{"name":"School of Economics and Management and Center for Economics and Finance (cef.UP) University of Porto  Porto Portugal"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3357-1195","authenticated-orcid":false,"given":"Jo\u00e3o","family":"Gama","sequence":"additional","affiliation":[{"name":"INESCTEC Campus da Faculdade de Engenharia da Universidade do Porto  Porto Portugal"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,2,1]]},"reference":[{"key":"e_1_2_11_2_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292500.3330701"},{"key":"e_1_2_11_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-4076(92)90066-Z"},{"key":"e_1_2_11_4_1","doi-asserted-by":"publisher","DOI":"10.2307\/2109358"},{"key":"e_1_2_11_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2018.05.004"},{"key":"e_1_2_11_6_1","doi-asserted-by":"publisher","DOI":"10.1002\/jae.1152"},{"key":"e_1_2_11_7_1","doi-asserted-by":"publisher","DOI":"10.1080\/713665670"},{"key":"e_1_2_11_8_1","doi-asserted-by":"publisher","DOI":"10.1093\/jjfinec\/nbp001"},{"key":"e_1_2_11_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbankfin.2022.106426"},{"key":"e_1_2_11_10_1","doi-asserted-by":"publisher","DOI":"10.1093\/jjfinec\/nby033"},{"issue":"4","key":"e_1_2_11_11_1","first-page":"40","article-title":"Improved Tracking\u2010Error Management for Active and Passive Investing","volume":"51","author":"De Nard G.","year":"2025","journal-title":"Journal of Portfolio Management"},{"key":"e_1_2_11_12_1","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.1080.0986"},{"key":"e_1_2_11_13_1","first-page":"1","article-title":"Statistical Comparisons of Classifiers Over Multiple Data Sets","volume":"7","author":"Demsar J.","year":"2006","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_11_14_1","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.1995.10524599"},{"key":"e_1_2_11_15_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1540-6261.1973.tb01451.x"},{"key":"e_1_2_11_16_1","doi-asserted-by":"publisher","DOI":"10.1198\/073500102288618487"},{"key":"e_1_2_11_17_1","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.2017.1345683"},{"key":"e_1_2_11_18_1","doi-asserted-by":"publisher","DOI":"10.3386\/w8554"},{"key":"e_1_2_11_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-405X(93)90023-5"},{"key":"e_1_2_11_20_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214506000001437"},{"key":"e_1_2_11_21_1","doi-asserted-by":"publisher","DOI":"10.1093\/rfs\/hhaa009"},{"key":"e_1_2_11_22_1","doi-asserted-by":"publisher","DOI":"10.2307\/2527081"},{"key":"e_1_2_11_23_1","doi-asserted-by":"publisher","DOI":"10.1111\/1468-0297.00152"},{"key":"e_1_2_11_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0169-2070(96)00719-4"},{"key":"e_1_2_11_25_1","volume-title":"Neural Computation","author":"Hochreiter S.","year":"1997"},{"key":"e_1_2_11_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2016.03.006"},{"key":"e_1_2_11_27_1","unstructured":"Jain S. andB. C.Wallace.2019. \u201cAttention Is Not Explanation.\u201d Preprint arXiv May 8.https:\/\/doi.org\/10.48550\/arXiv.1902.10186."},{"key":"e_1_2_11_28_1","volume-title":"Symmetrizing the Kullback\u2013Leibler Distance","author":"Johnson D. H.","year":"2001"},{"key":"e_1_2_11_29_1","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/acc638"},{"key":"e_1_2_11_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jempfin.2014.06.005"},{"key":"e_1_2_11_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbankfin.2012.05.005"},{"key":"e_1_2_11_32_1","doi-asserted-by":"publisher","DOI":"10.1142\/S0219024900000255"},{"key":"e_1_2_11_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_11_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0927-5398(03)00007-0"},{"key":"e_1_2_11_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0047-259X(03)00096-4"},{"key":"e_1_2_11_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmva.2015.04.006"},{"key":"e_1_2_11_37_1","doi-asserted-by":"publisher","DOI":"10.1214\/19-AOS1921"},{"key":"e_1_2_11_38_1","doi-asserted-by":"publisher","DOI":"10.1093\/jjfinec\/nbaa007"},{"key":"e_1_2_11_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2019.01.078"},{"key":"e_1_2_11_40_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1540-6261.1952.tb01525.x"},{"key":"e_1_2_11_41_1","doi-asserted-by":"publisher","DOI":"10.2469\/faj.v45.n1.31"},{"key":"e_1_2_11_42_1","volume-title":"RiskMetrics\u2122\u2014Technical Document","author":"Morgan J. P.","year":"1996"},{"key":"e_1_2_11_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.119640"},{"key":"e_1_2_11_44_1","doi-asserted-by":"publisher","DOI":"10.2307\/1913610"},{"key":"e_1_2_11_45_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-022-10226-0"},{"key":"e_1_2_11_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2006.03.010"},{"key":"e_1_2_11_47_1","volume-title":"The Role of Deep Learning in Financial Asset Management: A Systematic Review","author":"Reis P.","year":"2024"},{"key":"e_1_2_11_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.physa.2011.07.023"},{"key":"e_1_2_11_49_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbankfin.2018.08.013"},{"key":"e_1_2_11_50_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-024-09574-5"},{"key":"e_1_2_11_51_1","volume-title":"Advances in Neural Information Processing Systems","author":"Vaswani A.","year":"2017"},{"key":"e_1_2_11_52_1","unstructured":"Wiegreffe S. andY.Pinter.2019. \u201cAttention Is Not Not explanation.\u201d Preprint arXiv September 5.https:\/\/doi.org\/10.48550\/arXiv.1908.04626."},{"key":"e_1_2_11_53_1","doi-asserted-by":"publisher","DOI":"10.1093\/jjfinec\/nbae026"}],"container-title":["Journal of Forecasting"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/for.70110","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/for.70110","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/for.70110","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T02:59:21Z","timestamp":1780455561000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/for.70110"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,1]]},"references-count":52,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1002\/for.70110"],"URL":"https:\/\/doi.org\/10.1002\/for.70110","archive":["Portico"],"relation":{},"ISSN":["0277-6693","1099-131X"],"issn-type":[{"value":"0277-6693","type":"print"},{"value":"1099-131X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,1]]},"assertion":[{"value":"2025-06-02","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-13","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-01","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}