A Multi-Model Hybrid Framework for Forecasting International Crude Oil Benchmarks

A Multi-Model Hybrid Framework for Forecasting International Crude Oil Benchmarks | 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 Research Article A Multi-Model Hybrid Framework for Forecasting International Crude Oil Benchmarks Babafemi Daniel Ogunbona, Semiu Ayinla Alayande This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8647765/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 crude oil prices is essential for energy policy, investment decisions, and economic planning, yet remains challenging due to pronounced nonlinearity and volatility in oil markets. While hybrid models combining statistical and machine learning techniques are prevalent, the individual contribution of their components is seldom examined. This study presents a comprehensive comparison of standalone and hybrid machine learning models for forecasting monthly prices of three major crude oil benchmarks—Bonny Light, Brent, and West Texas Intermediate (WTI)—using monthly data from January 1998 to July 2025. The models considered include ARIMA, long short-term memory (LSTM) networks, Extreme Gradient Boosting (XGBoost), multilayer perceptrons (MLP), simple average hybrids, and multi-stage stacked hybrid architectures. Results confirm the dominance of XGBoost as the best standalone model. Simple averaging hybrids offer limited gains, whereas stacked hybrid models achieve consistent performance improvements. The proposed ARIMA–LSTM–XGBoost–MLP framework achieves the best performance, yielding RMSE values of 4.216 (Bonny), 3.881 (Brent), and 2.984 (WTI) with R² exceeding 0.95. Crucially, SHAP and sensitivity analysis deconstruct this optimal hybrid, revealing that over 90% of its predictive contribution originates from the XGBoost prediction, while ARIMA and LSTM inputs contribute minimally. This finding challenges the narrative that complex hybridization inherently benefits from model diversity. Instead, it shows that ensemble superiority can be predominantly attributed to a single, powerful algorithm. Our work emphasizes the importance of interpretability in hybrid modeling and offers practical guidance for efficient forecasting system design. Econometrics Macroeconomics Artificial Intelligence and Machine Learning Applied Statistics Hybrid models Machine learning Model interpretability SHAP analysis Full Text Additional Declarations The authors declare no competing interests. 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-8647765","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":577377526,"identity":"4883c79f-088d-4ca4-9789-40a3a4c717f2","order_by":0,"name":"Babafemi Daniel Ogunbona","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBUlEQVRIiWNgGAWjYJACZhBhwJB8AMIFUTzEaUlLIFlLjgFxWvj515g9Lqi4I2/OnvPxw48Kmzy+4w2MD962MeTJO2DXIjnjjbnxjDPPDHf2vN0s2XMmrVjyzAFmw7ltDMWGB7BrMbhxxkyat+0w44YbuRukGdsOJ264kcAGFGFI3NiAX4v9hhs5j38z/gNrYf+NV8v5HrAWoMocNmnGBogtzCAt83F4X3IGW5n0jDOHk3f2PDOz7DmWljjzzMFmyTnnJBI34Ayxw9ukCyoO225nT35840eNTWLf8eaDH96U2STOx+EwBokEDCFGkFoJBoMDuKzBJcEgj8uWUTAKRsEoGGkAAK5gZpDV3vDvAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0006-2404-1909","institution":"Adeyemi Federal University of Education, Ondo, Nigeria","correspondingAuthor":true,"prefix":"","firstName":"Babafemi","middleName":"Daniel","lastName":"Ogunbona","suffix":""},{"id":577377695,"identity":"1cebdeb7-0f50-4325-b427-260b67eacc5f","order_by":1,"name":"Semiu Ayinla Alayande","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Semiu","middleName":"Ayinla","lastName":"Alayande","suffix":""}],"badges":[],"createdAt":"2026-01-20 10:14:17","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-8647765/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8647765/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100785442,"identity":"1d9431ab-1a23-4b59-a3cd-a1407c421af6","added_by":"auto","created_at":"2026-01-21 11:56:01","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":379392,"visible":true,"origin":"","legend":"","description":"","filename":"AMultiModelHybridFrameworkforForecastingInternationalCrudeOilBenchmarks.docx","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/f4dfd0012170934574817a06.docx"},{"id":100785254,"identity":"3858d885-d62e-45e1-b31a-6a45280e6228","added_by":"auto","created_at":"2026-01-21 11:55:26","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs8647765.json","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/c82fc0d5b8d234a842cae2f4.json"},{"id":100785441,"identity":"952fee3b-c719-485a-8a71-ed3692d5bfc4","added_by":"auto","created_at":"2026-01-21 11:56:01","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103271,"visible":true,"origin":"","legend":"","description":"","filename":"rs86477650enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/8e360214822163c547146b38.xml"},{"id":100785275,"identity":"d4d5d22a-e3b0-4947-9967-9c6922206d0c","added_by":"auto","created_at":"2026-01-21 11:55:35","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67122,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/3169f3c127b492011f4b2c83.png"},{"id":100785216,"identity":"32a67810-393d-4e9b-80e2-dd2e82022295","added_by":"auto","created_at":"2026-01-21 11:55:23","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74112,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/2993a850677d95b637f1474c.png"},{"id":100785428,"identity":"0c27271b-f57b-4ef5-b0d6-0cb7ff9b08ff","added_by":"auto","created_at":"2026-01-21 11:55:48","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162770,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/c665454a934ea31340984960.png"},{"id":100785277,"identity":"0528c323-21a9-46aa-bd16-75d6019acdbe","added_by":"auto","created_at":"2026-01-21 11:55:35","extension":"wmf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3080,"visible":true,"origin":"","legend":"","description":"","filename":"image1.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/8deb15b264d3a4a140ce96ce.wmf"},{"id":100785257,"identity":"9e55ec35-b445-4a4c-9224-10cebf5e7c58","added_by":"auto","created_at":"2026-01-21 11:55:27","extension":"wmf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2572,"visible":true,"origin":"","legend":"","description":"","filename":"image2.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/b9de5078c6800f1f59afbf64.wmf"},{"id":100785443,"identity":"f7ec0216-55b3-457b-bb67-405f30ffc4cb","added_by":"auto","created_at":"2026-01-21 11:56:01","extension":"wmf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2906,"visible":true,"origin":"","legend":"","description":"","filename":"image3.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/04c06458f826780910792844.wmf"},{"id":100785394,"identity":"7e9da101-58ed-4b43-96c2-8246f3e51021","added_by":"auto","created_at":"2026-01-21 11:55:45","extension":"wmf","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3534,"visible":true,"origin":"","legend":"","description":"","filename":"image4.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/29bbbfbb027677167aad4558.wmf"},{"id":100785276,"identity":"0ee08681-90a0-405e-9b7f-6523c6249810","added_by":"auto","created_at":"2026-01-21 11:55:35","extension":"wmf","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":716,"visible":true,"origin":"","legend":"","description":"","filename":"image5.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/bfbebabc7a811bceb28cafb0.wmf"},{"id":100785375,"identity":"4b0e042d-ca4c-4ccd-b012-b08ef700c61d","added_by":"auto","created_at":"2026-01-21 11:55:39","extension":"wmf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":830,"visible":true,"origin":"","legend":"","description":"","filename":"image6.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/56de251c8106f0ac55d5f33a.wmf"},{"id":100785449,"identity":"cc32110a-696e-403a-ad2b-c2f2c4017955","added_by":"auto","created_at":"2026-01-21 11:56:04","extension":"wmf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":792,"visible":true,"origin":"","legend":"","description":"","filename":"image7.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/bfa1e6821c7b452496ff385d.wmf"},{"id":100785378,"identity":"04ab210c-9eb4-4e23-819b-604a548773b7","added_by":"auto","created_at":"2026-01-21 11:55:40","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29900,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/fd00ca14222e3404237392d4.png"},{"id":100785451,"identity":"618d93d3-39ee-4a4f-8a21-429673012ba1","added_by":"auto","created_at":"2026-01-21 11:56:06","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43879,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/753e04609c5a97e8df4976ab.png"},{"id":100785250,"identity":"3259b221-59e9-4a13-b783-fe52c4f691c8","added_by":"auto","created_at":"2026-01-21 11:55:24","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37275,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/115fc77be0ea8fc9cf81792c.png"},{"id":100785261,"identity":"b01df92b-7fd1-4274-991c-6a638a4fb8a6","added_by":"auto","created_at":"2026-01-21 11:55:28","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2025,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/52492f2b7bf0254f58fdfacd.png"},{"id":100796832,"identity":"45d32609-88f1-4543-9df7-b9cf915dd135","added_by":"auto","created_at":"2026-01-21 13:46:16","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1628,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/64f74c5e3dd08707f2ca6f3d.png"},{"id":100785369,"identity":"52974822-cdfb-425e-8767-7a5ff0cae82e","added_by":"auto","created_at":"2026-01-21 11:55:38","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2192,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/e3dba1ee007b87049e64ea08.png"},{"id":100785490,"identity":"99059f3c-83e1-4c0a-8aa4-8866049810eb","added_by":"auto","created_at":"2026-01-21 11:56:16","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1705,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/7ab4a3591d719ed894399224.png"},{"id":100785429,"identity":"2b794dc6-c8d2-4058-93ff-f2aab98e54cd","added_by":"auto","created_at":"2026-01-21 11:55:50","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":281,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/85c35bb410371f55115571f0.png"},{"id":100785487,"identity":"7bb64035-3800-4dd7-ae7b-e0173eb3bf8b","added_by":"auto","created_at":"2026-01-21 11:56:14","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":313,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/f33e63d9f9fd6560f3a0726d.png"},{"id":100785062,"identity":"b30612a8-d055-4d5e-ac89-bb1b46717723","added_by":"auto","created_at":"2026-01-21 11:55:00","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":292,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/cd2738f3a3f82713e2bee8b7.png"},{"id":100785130,"identity":"e68a858b-3245-43ba-a449-0699f7b2a71b","added_by":"auto","created_at":"2026-01-21 11:55:08","extension":"xml","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102441,"visible":true,"origin":"","legend":"","description":"","filename":"rs86477650structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/1ad5ca650f26ef03a3ebf91a.xml"},{"id":100785072,"identity":"15d5115e-082b-47cc-95eb-af153a364e07","added_by":"auto","created_at":"2026-01-21 11:55:04","extension":"html","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109438,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1/82fc617331774b5c51aeced9.html"},{"id":100798129,"identity":"e5400c67-ef71-4b8d-8595-2aae9fdf1eae","added_by":"auto","created_at":"2026-01-21 13:52:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":545370,"visible":true,"origin":"","legend":"","description":"","filename":"AMultiModelHybridFrameworkforForecastingInternationalCrudeOilBenchmarks.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8647765/v1_covered_10176654-d754-4b9e-9c9d-c706a6c51704.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eA Multi-Model Hybrid Framework for Forecasting International Crude Oil Benchmarks\u003c/strong\u003e\u003c/p\u003e","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":"Hybrid models, Machine learning, Model interpretability, SHAP analysis","lastPublishedDoi":"10.21203/rs.3.rs-8647765/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8647765/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccurate forecasting of crude oil prices is essential for energy policy, investment decisions, and economic planning, yet remains challenging due to pronounced nonlinearity and volatility in oil markets. While hybrid models combining statistical and machine learning techniques are prevalent, the individual contribution of their components is seldom examined. This study presents a comprehensive comparison of standalone and hybrid machine learning models for forecasting monthly prices of three major crude oil benchmarks\u0026mdash;Bonny Light, Brent, and West Texas Intermediate (WTI)\u0026mdash;using monthly data from January 1998 to July 2025. The models considered include ARIMA, long short-term memory (LSTM) networks, Extreme Gradient Boosting (XGBoost), multilayer perceptrons (MLP), simple average hybrids, and multi-stage stacked hybrid architectures. Results confirm the dominance of XGBoost as the best standalone model. Simple averaging hybrids offer limited gains, whereas stacked hybrid models achieve consistent performance improvements. The proposed ARIMA\u0026ndash;LSTM\u0026ndash;XGBoost\u0026ndash;MLP framework achieves the best performance, yielding RMSE values of 4.216 (Bonny), 3.881 (Brent), and 2.984 (WTI) with R\u0026sup2; exceeding 0.95. Crucially, SHAP and sensitivity analysis deconstruct this optimal hybrid, revealing that over 90% of its predictive contribution originates from the XGBoost prediction, while ARIMA and LSTM inputs contribute minimally. This finding challenges the narrative that complex hybridization inherently benefits from model diversity. Instead, it shows that ensemble superiority can be predominantly attributed to a single, powerful algorithm. Our work emphasizes the importance of interpretability in hybrid modeling and offers practical guidance for efficient forecasting system design.\u003c/p\u003e","manuscriptTitle":"A Multi-Model Hybrid Framework for Forecasting International Crude Oil Benchmarks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-21 11:03:43","doi":"10.21203/rs.3.rs-8647765/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":"860196bf-353e-4e91-b30e-4cb525d85022","owner":[],"postedDate":"January 21st, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":61429540,"name":"Econometrics"},{"id":61429541,"name":"Macroeconomics"},{"id":61429542,"name":"Artificial Intelligence and Machine Learning"},{"id":61429543,"name":"Applied Statistics"}],"tags":[],"updatedAt":"2026-01-21T11:03:43+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-21 11:03:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8647765","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8647765","identity":"rs-8647765","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}