Explaining Black-Box Models Through Statistical Inference

Explaining Black-Box Models Through Statistical Inference | 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 Explaining Black-Box Models Through Statistical Inference Arsh Chowdhry, Umair Rehman, Apurva Narayan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7830468/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 Explainable AI for tabular models underpins decisions in finance, healthcare, and policy, yet today’s explanations are dominated by heuristics without statistical guarantees. We introduce Stat-XAI, a model-agnostic framework that converts explanations into testable statistical statements. For each feature, Stat-XAI assesses association with model predictions on held-out data via appropriate hypothesis tests and reports standardized effect sizes (e.g., $\eta^2$, $R^2$, Cramér's $V$), yielding compact, uncertainty-aware rankings. Across six synthetic datasets with known causal structure and two real benchmarks, Stat-XAI delivers stable, parsimonious attributions, filters spurious correlates, and achieves orders-of-magnitude lower runtime than SHAP while maintaining faithfulness. We quantify stability under perturbations and show that interaction testing clarifies when pairwise dependencies meaningfully alter importance. By elevating explanation from heuristic scoring to inferential analysis, Stat-XAI provides a rigorous, reproducible pathway for trustworthy tabular AI—supporting scrutiny, governance, and human decision-making where reliability matters most. Physical sciences/Mathematics and computing/Computer science Physical sciences/Mathematics and computing/Statistics Explainable AI Inferential Statistics Interpretability Machine Learning Full Text Additional Declarations There is NO Competing Interest. Supplementary Files NMISupplementaryDocument.pdf Explaining Black-Box Models Through Statistical Inference. 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-7830468","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":528714124,"identity":"a64129eb-436e-4790-bd03-afc9e2a35942","order_by":0,"name":"Arsh Chowdhry","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYHACNiA+AMQ5QFwBxMzMDaRoOQPSwkiKFsY2kAABLfyzjz97wFBzR87geO6xz7zzaqP524FaflRsw6lF4lyOuQHDsWfGBmfeJc/m3XY8d8ZhxgbGnjO3cVtzhodNgrHhcOK2GznGzLzbjuU2ALUwM7bh1iJ/hv0ZkpY5x3LnE9JicIbBDElLQ03uBkJaDM/wmEkkHDtsbH/mjTHjnGMHcjcCtRzE5xc5kMM+1ByWk2zPMWZ4U1OXO+/84YMPflTg8T4IJEBpJh6Gw2DGAfzqkQDjD4Y6ohWPglEwCkbByAEArXtdJ9s2yXkAAAAASUVORK5CYII=","orcid":"","institution":"University of Western Ontario","correspondingAuthor":true,"prefix":"","firstName":"Arsh","middleName":"","lastName":"Chowdhry","suffix":""},{"id":528714125,"identity":"0ecb3bc8-0373-4d03-a962-e6a86983bd09","order_by":1,"name":"Umair Rehman","email":"","orcid":"","institution":"University of Western Ontario","correspondingAuthor":false,"prefix":"","firstName":"Umair","middleName":"","lastName":"Rehman","suffix":""},{"id":528714126,"identity":"10053bd5-5ab4-4215-a2a1-ad6d1450469a","order_by":2,"name":"Apurva Narayan","email":"","orcid":"","institution":"University of Western Ontario","correspondingAuthor":false,"prefix":"","firstName":"Apurva","middleName":"","lastName":"Narayan","suffix":""}],"badges":[],"createdAt":"2025-10-10 23:40:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7830468/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7830468/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94109431,"identity":"286a61f0-0852-45ff-be45-3e478a16407f","added_by":"auto","created_at":"2025-10-22 13:03:49","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3368166,"visible":true,"origin":"","legend":"Article File","description":"","filename":"NatureMachineIntelligencePaper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7830468/v1_covered_af846974-9a57-4943-b9f2-f86dc00f1a10.pdf"},{"id":93981483,"identity":"8e1a631a-b851-430d-9027-38b67fd6250d","added_by":"auto","created_at":"2025-10-21 02:55:44","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4780199,"visible":true,"origin":"","legend":"Explaining Black-Box Models Through Statistical Inference.","description":"","filename":"NMISupplementaryDocument.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7830468/v1/646056905e83949287afccfe.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Explaining Black-Box Models Through Statistical Inference","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":"Explainable AI, Inferential Statistics, Interpretability, Machine Learning","lastPublishedDoi":"10.21203/rs.3.rs-7830468/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7830468/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Explainable AI for tabular models underpins decisions in finance, healthcare, and policy, yet today’s explanations are dominated by heuristics without statistical guarantees. We introduce Stat-XAI, a model-agnostic framework that converts explanations into testable statistical statements. For each feature, Stat-XAI assesses association with model predictions on held-out data via appropriate hypothesis tests and reports standardized effect sizes (e.g., $\\eta^2$, $R^2$, Cramér's $V$), yielding compact, uncertainty-aware rankings. Across six synthetic datasets with known causal structure and two real benchmarks, Stat-XAI delivers stable, parsimonious attributions, filters spurious correlates, and achieves orders-of-magnitude lower runtime than SHAP while maintaining faithfulness. We quantify stability under perturbations and show that interaction testing clarifies when pairwise dependencies meaningfully alter importance. By elevating explanation from heuristic scoring to inferential analysis, Stat-XAI provides a rigorous, reproducible pathway for trustworthy tabular AI—supporting scrutiny, governance, and human decision-making where reliability matters most.","manuscriptTitle":"Explaining Black-Box Models Through Statistical Inference","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-21 02:55:39","doi":"10.21203/rs.3.rs-7830468/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":"3349ef06-447e-4150-8647-dd5e07f831d4","owner":[],"postedDate":"October 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56191450,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":56191451,"name":"Physical sciences/Mathematics and computing/Statistics"}],"tags":[],"updatedAt":"2025-10-22T12:55:40+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-21 02:55:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7830468","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7830468","identity":"rs-7830468","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}