Research on Transfer Learning and Algorithm Fairness Calibration in Cross-Market Credit Scoring

Research on Transfer Learning and Algorithm Fairness Calibration in Cross-Market Credit Scoring | 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 Research on Transfer Learning and Algorithm Fairness Calibration in Cross-Market Credit Scoring Jiajing Wang, Yiran Xiao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7927361/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 When credit scoring models are applied across markets, their transfer performance often deteriorates due to borrower feature distribution and sample heterogeneity. This paper systematically evaluates the transfer effectiveness of cross-market credit scoring models using 2 million credit card account records from China and the United States. It proposes a fairness calibration method combining domain-adaptive weighting with monotonic constraints. Experimental results show that directly transferred models experience a 7.2 percentage point decline in AUC in the target market. After applying the proposed method, performance loss is reduced to 1.3 percentage points while group fairness metrics (equity gap) improve by 41%. The study demonstrates that integrating transfer learning with fairness constraints not only enhances the robustness and generalizability of cross-domain scoring models but also ensures credit fairness across different groups. This provides an effective pathway for applying data science in cross-border credit accessibility and compliance modeling. Artificial Intelligence and Machine Learning Theoretical Computer Science Financial Mathematics transfer learning credit scoring fairness calibration cross-domain modeling consumer finance algorithmic fairness 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-7927361","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":533663402,"identity":"ad431342-3826-4881-88ef-24169abe6119","order_by":0,"name":"Jiajing Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYBACPmYGNhDNIy///OADEIOPkBY2qBYZw4acZAOQFjaCWhggWmwYDiSYSUBFCGhhZ3724OMOBh7GhgNplV9z7GTYGJgfPrqB12Fs5oYzzzDwsDM2Hrstuy0Z6DA2Y+McvFp42KR524C2NDOk3ZbcBuQCvSNNUMtfoBaGYwxmxZLb6onUwgjScobBjPHjtsPEaGEzk+xtk+AxnMGTLM247TgPUAS/X/j5Dz+T+NlmYy8vwX7w489t1fb87M0PH+PTAgXgGGFg5gGThJUjAOMPUlSPglEwCkbBiAEADfY2f2ya3vEAAAAASUVORK5CYII=","orcid":"","institution":"Columbia University, New York, US, 10027","correspondingAuthor":true,"prefix":"","firstName":"","middleName":"","lastName":"Jiajing Wang","suffix":""},{"id":533663403,"identity":"8e9f1add-6069-47d1-9588-5ba74595abc3","order_by":1,"name":"Yiran Xiao","email":"","orcid":"","institution":"Northwestern University, Evanston, US, 60208","correspondingAuthor":false,"prefix":"","firstName":"","middleName":"","lastName":"Yiran Xiao","suffix":""}],"badges":[],"createdAt":"2025-10-23 01:59:39","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7927361/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7927361/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94245569,"identity":"4a245e18-ed7f-41eb-85fb-6a4076b0e9e5","added_by":"auto","created_at":"2025-10-24 05:23:49","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":412507,"visible":true,"origin":"","legend":"","description":"","filename":"10.20ICDSIC20254zz35144.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7927361/v1_covered_5aea8fed-e3cc-42b4-a018-a63a01bec830.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eResearch on Transfer Learning and Algorithm Fairness Calibration in Cross-Market Credit Scoring\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":"transfer learning, credit scoring, fairness calibration, cross-domain modeling, consumer finance, algorithmic fairness","lastPublishedDoi":"10.21203/rs.3.rs-7927361/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7927361/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWhen credit scoring models are applied across markets, their transfer performance often deteriorates due to borrower feature distribution and sample heterogeneity. This paper systematically evaluates the transfer effectiveness of cross-market credit scoring models using 2 million credit card account records from China and the United States. It proposes a fairness calibration method combining domain-adaptive weighting with monotonic constraints. Experimental results show that directly transferred models experience a 7.2 percentage point decline in AUC in the target market. After applying the proposed method, performance loss is reduced to 1.3 percentage points while group fairness metrics (equity gap) improve by 41%. The study demonstrates that integrating transfer learning with fairness constraints not only enhances the robustness and generalizability of cross-domain scoring models but also ensures credit fairness across different groups. This provides an effective pathway for applying data science in cross-border credit accessibility and compliance modeling.\u003c/p\u003e","manuscriptTitle":"Research on Transfer Learning and Algorithm Fairness Calibration in Cross-Market Credit Scoring","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-24 05:23:14","doi":"10.21203/rs.3.rs-7927361/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":"9d7cc715-defc-428b-83e0-f91e89f5bc4d","owner":[],"postedDate":"October 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56738528,"name":"Artificial Intelligence and Machine Learning"},{"id":56738529,"name":"Theoretical Computer Science"},{"id":56738530,"name":"Financial Mathematics"}],"tags":[],"updatedAt":"2025-10-24T05:23:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-24 05:23:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7927361","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7927361","identity":"rs-7927361","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}