Decomposition-Enhanced Network for Financial Time Series Forecasting

Decomposition-Enhanced Network for Financial Time Series Forecasting | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Decomposition-Enhanced Network for Financial Time Series Forecasting Jinyuan Huang, Qianqian Sun, Xinghua Zhang, Hong Du, Zheng Xu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9180493/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract The extreme non-stationarity, high noise levels, and multi-timescale coupling in financial futures markets pose major challenges for time series forecasting. Existing models often struggle to disentangle localized shocks from global trends due to incompatible inductive biases. To address this issue, we propose a Decomposition-Enhanced Network (DENet). Following a divide-and-conquer paradigm, DENet adopts a multi-stream architecture: the main path extracts stable trends via moving averages and dual-path linear projections, while Auxiliary Stream I captures seasonal and local cyclical patterns using depthwise separable convolutions, and Auxiliary Stream II models high-frequency dynamics through a nonlinear autoregressive-style mapping. These components are integrated via an adaptive fusion mechanism, balancing global robustness and local structural sensitivity. Experiments on real-world futures data demonstrate that DENet outperforms a wide range of state-of-the-art benchmarks. Compared with DLinear, PatchTST, TSMixer, and iTransformer, DENet achieves an average reduction of 12.84% in RMSE for daily forecasting and 9.35% in MAE over the 5-minute, 12-step forecasting horizon on iron ore futures. Furthermore, we integrate DENet's dual-scale predictions into the R-Breaker strategy with parameter switching and dynamic position sizing. Backtesting results show that the annualized return for silicomanganese futures increases by 121.16 percentage points over the baseline strategy. Ultimately, DENet effectively bridges advanced structural modeling and actionable algorithmic trading. Physical sciences/Mathematics and computing Physical sciences/Physics Decomposition-Enhanced Network(DENet) futures trading time series forecasting divide-and-conquer deep learning R-breaker trading strategy Full Text Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterial.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 15 Apr, 2026 Reviews received at journal 13 Apr, 2026 Reviewers agreed at journal 09 Apr, 2026 Reviewers agreed at journal 08 Apr, 2026 Reviews received at journal 06 Apr, 2026 Reviewers agreed at journal 06 Apr, 2026 Reviewers invited by journal 02 Apr, 2026 Editor invited by journal 25 Mar, 2026 Editor assigned by journal 23 Mar, 2026 Submission checks completed at journal 23 Mar, 2026 First submitted to journal 20 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9180493","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":618214232,"identity":"14ab6051-62e5-4669-b9ee-b3280d05b446","order_by":0,"name":"Jinyuan Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYBACNvb+jw8SKiSY+eUPJAIZNYS18PEcMDb4cMaGXXIGw2ODB2eOEdYiJ5FgJjmzLY3f4AbjM8mHLcxEOIznQJo0D9thaYbbzWkViQ1sDPzt3QkE/NJw2JqH57Ax45xjaTcSd8gwSJw5u4GALQcbb/NIHE5mZsgBajnDxmAgkUtAi0QygzSPweH6Nob8bwWJbczEaEljkpyRkMbMI5GQxkCcFp4zzAYfDtgwS/AcSJZIOHOMh6Bf5Nt7GB8k/pNgtj/ekPjxR0WNHH97L34tGICHNOWjYBSMglEwCrACAMhDSr18MnhhAAAAAElFTkSuQmCC","orcid":"","institution":"Henan Polytechnic University","correspondingAuthor":true,"prefix":"","firstName":"Jinyuan","middleName":"","lastName":"Huang","suffix":""},{"id":618214233,"identity":"f1b05c2b-9af1-4586-9bd8-5e5b78de1fd9","order_by":1,"name":"Qianqian Sun","email":"","orcid":"","institution":"Henan Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Qianqian","middleName":"","lastName":"Sun","suffix":""},{"id":618214234,"identity":"0ea8b423-a9a2-49b5-b8aa-c7bd517afe99","order_by":2,"name":"Xinghua Zhang","email":"","orcid":"","institution":"Henan Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Xinghua","middleName":"","lastName":"Zhang","suffix":""},{"id":618214235,"identity":"9f81f9f7-4235-42c9-97fd-71dffa59bbf8","order_by":3,"name":"Hong Du","email":"","orcid":"","institution":"Henan Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Du","suffix":""},{"id":618214236,"identity":"af8be39f-46c5-4a0c-94c8-e67d8d641caa","order_by":4,"name":"Zheng Xu","email":"","orcid":"","institution":"Henan Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2026-03-20 15:53:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9180493/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9180493/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106404565,"identity":"a2ffddca-0e6b-4047-aa8d-c4cf21a9fc5b","added_by":"auto","created_at":"2026-04-08 09:16:15","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1127789,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9180493/v1_covered_ef1b15e9-98a7-45d9-8415-b5c93a1f72f8.pdf"},{"id":106375733,"identity":"d7a4731e-6576-4345-b329-32ada420fd36","added_by":"auto","created_at":"2026-04-08 03:28:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2597334,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9180493/v1/81c823cd52c9637121f793c2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Decomposition-Enhanced Network for Financial Time Series Forecasting","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Decomposition-Enhanced Network(DENet), futures trading, time series forecasting, divide-and-conquer, deep learning, R-breaker trading strategy","lastPublishedDoi":"10.21203/rs.3.rs-9180493/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9180493/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The extreme non-stationarity, high noise levels, and multi-timescale coupling in financial futures markets pose major challenges for time series forecasting. Existing models often struggle to disentangle localized shocks from global trends due to incompatible inductive biases. To address this issue, we propose a Decomposition-Enhanced Network (DENet). Following a divide-and-conquer paradigm, DENet adopts a multi-stream architecture: the main path extracts stable trends via moving averages and dual-path linear projections, while Auxiliary Stream I captures seasonal and local cyclical patterns using depthwise separable convolutions, and Auxiliary Stream II models high-frequency dynamics through a nonlinear autoregressive-style mapping. These components are integrated via an adaptive fusion mechanism, balancing global robustness and local structural sensitivity. Experiments on real-world futures data demonstrate that DENet outperforms a wide range of state-of-the-art benchmarks. Compared with DLinear, PatchTST, TSMixer, and iTransformer, DENet achieves an average reduction of 12.84% in RMSE for daily forecasting and 9.35% in MAE over the 5-minute, 12-step forecasting horizon on iron ore futures. Furthermore, we integrate DENet's dual-scale predictions into the R-Breaker strategy with parameter switching and dynamic position sizing. Backtesting results show that the annualized return for silicomanganese futures increases by 121.16 percentage points over the baseline strategy. Ultimately, DENet effectively bridges advanced structural modeling and actionable algorithmic trading.","manuscriptTitle":"Decomposition-Enhanced Network for Financial Time Series Forecasting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-08 03:27:56","doi":"10.21203/rs.3.rs-9180493/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-15T10:26:27+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-13T06:20:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"222411833944488996519002934552822939550","date":"2026-04-09T15:55:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"208918660803680743060640267466530751540","date":"2026-04-08T05:57:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-06T11:48:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"52031050279518487495628764373172318728","date":"2026-04-06T11:19:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-02T22:20:44+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-25T05:41:58+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-23T10:49:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-23T10:49:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-20T15:37:25+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"53d1b132-9107-4f49-950a-dbe152b0737a","owner":[],"postedDate":"April 8th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65783735,"name":"Physical sciences/Mathematics and computing"},{"id":65783736,"name":"Physical sciences/Physics"}],"tags":[],"updatedAt":"2026-04-30T07:23:23+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-08 03:27:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9180493","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9180493","identity":"rs-9180493","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}