Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation

Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation | 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 Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation Zhengfa Hu, Xin Yang, Wenping Jiang, Yeli Shi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7726466/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Accurate forecasting of financial volatility is critical for risk management and investment decision-making. This study proposes a novel three-stage hybrid model (C-WG-BL) integrating Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), and Bidirectional Long Short-Term Memory (BiLSTM) networks. First, CEEMDAN decomposes the original volatility series into multi-scale Intrinsic Mode Functions (IMFs). Next, WGAN-GP is applied to high-frequency IMFs containing major noise and microstructural fluctuations, generating high-quality synthetic data to expand the training set and enhance the model’s ability to capture complex patterns. Finally, BiLSTM forecasts all IMFs (augmented high-frequency and original low-frequency), and the results are integrated to reconstruct the final prediction. Empirical analysis using 1-minute high-frequency data from the SSE 50, CSI 300, CSI 500, and SSE Composite Index shows that C-WG-BL significantly outperforms mainstream deep learning and ablation models. For the SSE 50, it improves R2 by 5.38% and reduces MAPE from 19.197% to 7.432% (–11.76 percentage points) compared with the best baseline (C-BiLSTM). The model also maintains high accuracy under extreme conditions such as the early COVID-19 outbreak, and Diebold–Mariano tests confirm statistically significant error reductions. This study offers an efficient, robust, and generalizable solution for high-frequency financial volatility prediction. Physical sciences/Engineering Physical sciences/Mathematics and computing CEEMDAN WGAN-GP BiLSTM volatility prediction Diebold Mariano test Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-7726466","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":525352295,"identity":"90bb33c4-440f-416b-96c7-6d3ef1ba89e6","order_by":0,"name":"Zhengfa Hu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIiWNgGAWjYBACNvbmgw8SDCTk+CUgAowNhLTw8RxLNvhQYWEsOQPIO0CMFjmJHDPBGWcqEjfcIFYLG1ALM2+bhLHx7R7Dzx8YbGQ3HGB+9gCvFp5nZY+BWuTM7pwxljjAkGa84QCbuQFeLezJ241BtpjdyDEAajmcuOEAD5sEXi0MCWbSQC2Jm2fkGP84wPCfCC0cKWaSM85IJG4AegpoywEitEACWcJY4kZamcUZg2TjmYfZzPBqkW8HR2WdHP+M5M03KirsZPuONz/DqwUNgIKKmQT1o2AUjIJRMAqwAwCLC0rWbHfs0QAAAABJRU5ErkJggg==","orcid":"","institution":"Hubei University of Automotive Technology","correspondingAuthor":true,"prefix":"","firstName":"Zhengfa","middleName":"","lastName":"Hu","suffix":""},{"id":525352297,"identity":"14957ce9-9cba-42d6-95c5-17055a22ce31","order_by":1,"name":"Xin Yang","email":"","orcid":"","institution":"Hubei University of Automotive Technology","correspondingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Yang","suffix":""},{"id":525352298,"identity":"a480fa6e-bb54-4ee6-a0a1-78923a738a09","order_by":2,"name":"Wenping Jiang","email":"","orcid":"","institution":"Hubei University of Automotive Technology","correspondingAuthor":false,"prefix":"","firstName":"Wenping","middleName":"","lastName":"Jiang","suffix":""},{"id":525352299,"identity":"b330a0c5-f9e0-424d-8025-64c783eb77d5","order_by":3,"name":"Yeli Shi","email":"","orcid":"","institution":"Hubei University of Automotive Technology","correspondingAuthor":false,"prefix":"","firstName":"Yeli","middleName":"","lastName":"Shi","suffix":""}],"badges":[],"createdAt":"2025-09-27 06:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7726466/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7726466/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92950656,"identity":"5c204274-dab7-4733-b7a6-a492d04d3a3c","added_by":"auto","created_at":"2025-10-07 13:06:40","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5799,"visible":true,"origin":"","legend":"","description":"","filename":"45d32b4f73584b6d925f7c0e22f3898b.json","url":"https://assets-eu.researchsquare.com/files/rs-7726466/v1/a79528ac4c8f2775dd516b2d.json"},{"id":105975689,"identity":"83884470-9224-4219-998e-323b0876b113","added_by":"auto","created_at":"2026-04-02 05:11:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":758720,"visible":true,"origin":"","legend":"","description":"","filename":"Article.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7726466/v1_covered_9ea9b883-4913-417b-88da-af59191c94e3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"CEEMDAN, WGAN-GP, BiLSTM, volatility prediction, Diebold Mariano test","lastPublishedDoi":"10.21203/rs.3.rs-7726466/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7726466/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Accurate forecasting of financial volatility is critical for risk management and investment decision-making. This study proposes a novel three-stage hybrid model (C-WG-BL) integrating Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), and Bidirectional Long Short-Term Memory (BiLSTM) networks. First, CEEMDAN decomposes the original volatility series into multi-scale Intrinsic Mode Functions (IMFs). Next, WGAN-GP is applied to high-frequency IMFs containing major noise and microstructural fluctuations, generating high-quality synthetic data to expand the training set and enhance the model’s ability to capture complex patterns. Finally, BiLSTM forecasts all IMFs (augmented high-frequency and original low-frequency), and the results are integrated to reconstruct the final prediction. Empirical analysis using 1-minute high-frequency data from the SSE 50, CSI 300, CSI 500, and SSE Composite Index shows that C-WG-BL significantly outperforms mainstream deep learning and ablation models. For the SSE 50, it improves R2 by 5.38% and reduces MAPE from 19.197% to 7.432% (–11.76 percentage points) compared with the best baseline (C-BiLSTM). The model also maintains high accuracy under extreme conditions such as the early COVID-19 outbreak, and Diebold–Mariano tests confirm statistically significant error reductions. This study offers an efficient, robust, and generalizable solution for high-frequency financial volatility prediction.","manuscriptTitle":"Research on Deep Learning Financial Volatility Prediction Method Based on Signal Decomposition and Data Augmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-07 13:05:03","doi":"10.21203/rs.3.rs-7726466/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"84df6acc-5be3-421f-a5e9-849f3bae9b40","owner":[],"postedDate":"October 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55830122,"name":"Physical sciences/Engineering"},{"id":55830123,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-02T05:10:49+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-07 13:05:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7726466","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7726466","identity":"rs-7726466","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}