{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,10]],"date-time":"2026-10-10T00:49:50Z","timestamp":1791593390467,"version":"4.3.4"},"reference-count":59,"publisher":"Oxford University Press (OUP)","license":[{"start":{"date-parts":[[2026,4,27]],"date-time":"2026-04-27T00:00:00Z","timestamp":1777248000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"DOI":"10.13039\/501100001321","name":"National Research Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001321","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014188","name":"MSIT","doi-asserted-by":"publisher","award":["RS-2022-NR068754"],"award-info":[{"award-number":["RS-2022-NR068754"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014188","name":"MSIT","doi-asserted-by":"publisher","award":["RS-2024-00335008"],"award-info":[{"award-number":["RS-2024-00335008"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014188","name":"MSIT","doi-asserted-by":"publisher","award":["RS-2025-24534596"],"award-info":[{"award-number":["RS-2025-24534596"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014364","name":"Samsung Science and Technology Foundation","doi-asserted-by":"publisher","award":["SSTF-BA2502-01"],"award-info":[{"award-number":["SSTF-BA2502-01"]}],"id":[{"id":"10.13039\/501100014364","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000092","name":"National Library of Medicine","doi-asserted-by":"publisher","award":["#1R01LM013361-01A1"],"award-info":[{"award-number":["#1R01LM013361-01A1"]}],"id":[{"id":"10.13039\/100000092","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Causal effects are often characterized at the population level, which can mask important heterogeneity across latent subgroups. Since the subgroup structure is unknown, identifying and evaluating subgroup specific effects is substantially more challenging than standard population level analysis. We address this problem by proposing Causal k-Means Clustering, a framework that uses k-means clustering ideas to recover unknown subgroup structure from individual level causal contrasts. The problem differs fundamentally from ordinary clustering because the objects to be clustered are unknown counterfactual functions. We first study a simple plug in estimator that is readily implemented with standard algorithms and establish its rate of convergence. We then develop a bias-corrected estimator using semiparametric efficiency theory and double machine learning and show that it attains fast root-n rates and asymptotic normality in large nonparametric models. The proposed methods are especially useful for modern outcome wide studies with multiple treatment levels, and the framework extends naturally to clustering based on more general pseudo-outcomes, including partially observed outcomes and other unknown functionals. We study finite sample performance in simulations and illustrate the method with an application to mobile supported self-management for chronic low back pain.<\/jats:p>","DOI":"10.1093\/jrsssb\/qkag068","type":"journal-article","created":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T11:33:12Z","timestamp":1775561592000},"source":"Crossref","is-referenced-by-count":1,"title":["Causal K-means clustering"],"prefix":"10.1093","author":[{"given":"Kwangho","family":"Kim","sequence":"first","affiliation":[{"name":"Department of Statistics, Korea University , 145 Anam-ro, Seongbuk-gu, Seoul 02841 ,","place":["Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jisu","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Statistics, Seoul National University , Seoul ,","place":["Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Edward H","family":"Kennedy","sequence":"additional","affiliation":[{"name":"Department of Statistics and Data Science, Carnegie Mellon University , Pittsburgh, PA 15213 ,","place":["USA"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,4,27]]},"reference":[{"key":"2026042702520200300_qkag068-B1","author":"Arthur","year":"2007"},{"issue":"27","key":"2026042702520200300_qkag068-B2","doi-asserted-by":"publisher","first-page":"7353","DOI":"10.1073\/pnas.1510489113","article-title":"Recursive partitioning for heterogeneous causal effects","volume":"113","author":"Athey","year":"2016","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"2","key":"2026042702520200300_qkag068-B3","doi-asserted-by":"publisher","first-page":"608","DOI":"10.1214\/009053606000001217","article-title":"Fast learning rates for plug-in classifiers","volume":"35","author":"Audibert","year":"2007","journal-title":"Annals of Statistics"},{"issue":"2","key":"2026042702520200300_qkag068-B4","doi-asserted-by":"publisher","first-page":"781","DOI":"10.1109\/TIT.2007.913516","article-title":"On the performance of clustering in Hilbert spaces","volume":"54","author":"Biau","year":"2008","journal-title":"IEEE Transactions on Information Theory"},{"key":"2026042702520200300_qkag068-B5","first-page":"585","article-title":"Convergence properties of the k-means algorithms","volume":"7","author":"Bottou","year":"1994","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"5","key":"2026042702520200300_qkag068-B6","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1257\/aer.p20171038","article-title":"Double\/debiased\/Neyman machine learning of treatment effects","volume":"107","author":"Chernozhukov","year":"2017,","journal-title":"American Economic Review"},{"key":"2026042702520200300_qkag068-B7","author":"Chernozhukov","year":"2018"},{"issue":"1","key":"2026042702520200300_qkag068-B8","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1111\/j.1468-0262.2005.00570.x","article-title":"An IV model of quantile treatment effects","volume":"73","author":"Chernozhukov","year":"2005","journal-title":"Econometrica: Journal of the Econometric Society"},{"key":"2026042702520200300_qkag068-B9","volume-title":"A probabilistic theory of pattern recognition","author":"Devroye","year":"2013"},{"key":"2026042702520200300_qkag068-B10","author":"Ding","year":"2004"},{"issue":"24","key":"2026042702520200300_qkag068-B11","doi-asserted-by":"publisher","first-page":"2867","DOI":"10.1002\/sim.4322","article-title":"Subgroup identification from randomized clinical trial data","volume":"30","author":"Foster","year":"2011","journal-title":"Statistics in Medicine"},{"key":"2026042702520200300_qkag068-B12","volume-title":"Foundations of quantization for probability distributions","author":"Graf","year":"2007"},{"issue":"4","key":"2026042702520200300_qkag068-B13","doi-asserted-by":"publisher","first-page":"413","DOI":"10.1017\/pan.2017.15","article-title":"Estimating heterogeneous treatment effects and the effects of heterogeneous treatments with ensemble methods","volume":"25","author":"Grimmer","year":"2017","journal-title":"Political Analysis"},{"issue":"2","key":"2026042702520200300_qkag068-B14","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1177\/0049124111400041","article-title":"Group-based trajectory modeling extended to account for nonrandom participant attrition","volume":"40","author":"Haviland","year":"2011","journal-title":"Sociological Methods & Research"},{"key":"2026042702520200300_qkag068-B15","author":"Hayden","year":"2009"},{"issue":"1","key":"2026042702520200300_qkag068-B16","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1097\/NNR.0000000000000856","article-title":"Feasibility of a gamified mobile-based self-management intervention for individuals with nonspecific chronic lower back pain","volume":"75","author":"Hong","year":"2022","journal-title":"Nursing Research"},{"issue":"1","key":"2026042702520200300_qkag068-B17","doi-asserted-by":"publisher","first-page":"443","DOI":"10.1214\/12-AOAS593","article-title":"Estimating treatment effect heterogeneity in randomized program evaluation","volume":"7","author":"Imai","year":"2013","journal-title":"The Annals of Applied Statistics"},{"key":"2026042702520200300_qkag068-B18","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9781139025751","volume-title":"Causal inference in statistics, social, and biomedical sciences","author":"Imbens","year":"2015"},{"issue":"8","key":"2026042702520200300_qkag068-B19","doi-asserted-by":"publisher","first-page":"651","DOI":"10.1016\/j.patrec.2009.09.011","article-title":"Data clustering: 50 years beyond k-means","volume":"31","author":"Jain","year":"2010","journal-title":"Pattern Recognition Letters"},{"issue":"7","key":"2026042702520200300_qkag068-B20","doi-asserted-by":"publisher","first-page":"881","DOI":"10.1109\/TPAMI.2002.1017616","article-title":"An efficient k-means clustering algorithm: Analysis and implementation","volume":"24","author":"Kanungo","year":"2002","journal-title":"IEEE Transactions on Pattern Analysis & Machine Intelligence"},{"key":"2026042702520200300_qkag068-B21","first-page":"141","volume-title":"Semiparametric theory and empirical processes in causal inference","author":"Kennedy","year":"2016"},{"issue":"2","key":"2026042702520200300_qkag068-B22","doi-asserted-by":"publisher","first-page":"3008","DOI":"10.1214\/23-EJS2157","article-title":"Towards optimal doubly robust estimation of heterogeneous causal effects","volume":"17","author":"Kennedy","year":"2023","journal-title":"Electronic Journal of Statistics"},{"key":"2026042702520200300_qkag068-B23","doi-asserted-by":"publisher","first-page":"207","DOI":"10.1201\/9781003216223","article-title":"Semiparametric doubly robust targeted double machine learning: A review","author":"Kennedy","year":"2024","journal-title":"Handbook of statistical methods for precision medicine"},{"issue":"4","key":"2026042702520200300_qkag068-B24","doi-asserted-by":"publisher","first-page":"2008","DOI":"10.1214\/19-AOS1874","article-title":"Sharp instruments for classifying compliers and generalizing causal effects","volume":"48","author":"Kennedy","year":"2020","journal-title":"Annals of Statistics"},{"key":"2026042702520200300_qkag068-B25","author":"Kim","year":"2022,"},{"key":"2026042702520200300_qkag068-B26","author":"Kim","year":"2023,"},{"issue":"4","key":"2026042702520200300_qkag068-B27","doi-asserted-by":"publisher","first-page":"661","DOI":"10.1111\/j.0887-378X.2004.00327.x","article-title":"Evidence-based medicine, heterogeneity of treatment effects, and the trouble with averages","volume":"82","author":"Kravitz","year":"2004","journal-title":"The Milbank Quarterly"},{"issue":"12","key":"2026042702520200300_qkag068-B28","doi-asserted-by":"publisher","first-page":"6084","DOI":"10.1016\/j.csda.2006.12.012","article-title":"Clustering data with measurement errors","volume":"51","author":"Kumar","year":"2007","journal-title":"Computational Statistics & Data Analysis"},{"issue":"10","key":"2026042702520200300_qkag068-B29","doi-asserted-by":"publisher","first-page":"4156","DOI":"10.1073\/pnas.1804597116","article-title":"Metalearners for estimating heterogeneous treatment effects using machine learning","volume":"116","author":"K\u00fcnzel","year":"2019","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"2","key":"2026042702520200300_qkag068-B30","doi-asserted-by":"publisher","first-page":"4239","DOI":"10.1214\/18-EJS1500","article-title":"A notion of stability for k-means clustering","volume":"12","author":"Le Gouic","year":"2018","journal-title":"Electronic Journal of Statistics"},{"key":"2026042702520200300_qkag068-B31","doi-asserted-by":"crossref","DOI":"10.1017\/9781108684163","volume-title":"Mining of massive data sets","author":"Leskovec","year":"2020"},{"issue":"5","key":"2026042702520200300_qkag068-B32","doi-asserted-by":"publisher","first-page":"1508","DOI":"10.1093\/jrsssb\/qkaf028","article-title":"Covariate-assisted bounds on causal effects with instrumental variables","volume":"87","author":"Levis","year":"2025","journal-title":"Journal of the Royal Statistical Society: Series B, Statistical Methodology"},{"issue":"2","key":"2026042702520200300_qkag068-B33","doi-asserted-by":"publisher","first-page":"592","DOI":"10.1214\/14-AOS1293","article-title":"Nonasymptotic bounds for vector quantization in Hilbert spaces","volume":"43","author":"Levrard","year":"2015","journal-title":"Annals of Statistics"},{"issue":"1","key":"2026042702520200300_qkag068-B34","first-page":"1","article-title":"Quantization\/clustering: When and why does k-means work?","volume":"159","author":"Levrard","year":"2018","journal-title":"Journal de la Soci\u00e9t\u00e9 Fran\u00e7aise de Statistique"},{"issue":"6","key":"2026042702520200300_qkag068-B35","doi-asserted-by":"publisher","first-page":"1728","DOI":"10.1109\/18.340451","article-title":"Rates of convergence in the source coding theorem, in empirical quantizer design, and in universal lossy source coding","volume":"40","author":"Linder","year":"1994","journal-title":"IEEE Transactions on Information Theory"},{"issue":"2","key":"2026042702520200300_qkag068-B36","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1109\/TIT.1982.1056489","article-title":"Least squares quantization in PCM","volume":"28","author":"Lloyd","year":"1982","journal-title":"IEEE Transactions on Information Theory"},{"issue":"2","key":"2026042702520200300_qkag068-B37","doi-asserted-by":"publisher","first-page":"713","DOI":"10.1214\/15-AOS1384","article-title":"Statistical inference for the mean outcome under a possibly non-unique optimal treatment strategy","volume":"44","author":"Luedtke","year":"2016","journal-title":"Annals of Statistics"},{"issue":"17","key":"2026042702520200300_qkag068-B38","doi-asserted-by":"publisher","DOI":"10.1920\/wp.cem.2017.4117","article-title":"Cross-fitting and fast remainder rates for semiparametric estimation","volume":"41","author":"Newey","year":"2018","journal-title":"Institute for Fiscal Studies"},{"issue":"2","key":"2026042702520200300_qkag068-B39","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1093\/biomet\/asaa076","article-title":"Quasi-oracle estimation of heterogeneous treatment effects","volume":"108","author":"Nie","year":"2021","journal-title":"Biometrika"},{"issue":"1","key":"2026042702520200300_qkag068-B40","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1214\/aos\/1176345339","article-title":"Strong consistency of k-means clustering","volume":"9","author":"Pollard","year":"1981","journal-title":"Annals of Statistics"},{"issue":"4","key":"2026042702520200300_qkag068-B41","doi-asserted-by":"publisher","first-page":"919","DOI":"10.1214\/aop\/1176993713","article-title":"A central limit theorem for k-means clustering","volume":"10","author":"Pollard","year":"1982","journal-title":"Annals of Probability"},{"key":"2026042702520200300_qkag068-B42","first-page":"335","volume-title":"Higher order influence functions and minimax estimation of nonlinear functionals","author":"Robins","year":"2008"},{"issue":"5","key":"2026042702520200300_qkag068-B43","doi-asserted-by":"publisher","first-page":"688","DOI":"10.1037\/h0037350","article-title":"Estimating causal effects of treatments in randomized and nonrandomized studies","volume":"66","author":"Rubin","year":"1974","journal-title":"Journal of Educational Psychology"},{"issue":"4","key":"2026042702520200300_qkag068-B44","article-title":"Handling missing data in model-based clustering","volume":"1050","author":"Serafini","year":"2020","journal-title":"stat"},{"issue":"3","key":"2026042702520200300_qkag068-B45","doi-asserted-by":"publisher","first-page":"831","DOI":"10.1214\/16-BA1022","article-title":"Latent class mixture models of treatment effect heterogeneity","volume":"12","author":"Shahn","year":"2017","journal-title":"Bayesian Analysis"},{"key":"2026042702520200300_qkag068-B46","author":"Shalit","year":"2017"},{"issue":"4","key":"2026042702520200300_qkag068-B47","doi-asserted-by":"publisher","first-page":"2337","DOI":"10.5705\/ss.202017.0093","article-title":"Clustering in general measurement error models","volume":"28","author":"Su","year":"2018","journal-title":"Statistica Sinica"},{"issue":"3","key":"2026042702520200300_qkag068-B48","doi-asserted-by":"publisher","first-page":"323","DOI":"10.3102\/1076998620951983","article-title":"Hybridizing machine learning methods and finite mixture models for estimating heterogeneous treatment effects in latent classes","volume":"46","author":"Suk","year":"2021","journal-title":"Journal of Educational and Behavioral Statistics"},{"issue":"1","key":"2026042702520200300_qkag068-B49","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1111\/j.0006-341X.2005.031032.x","article-title":"Tight clustering: A resampling-based approach for identifying stable and tight patterns in data","volume":"61","author":"Tseng","year":"2005","journal-title":"Biometrics"},{"key":"2026042702520200300_qkag068-B50","volume-title":"Semiparametric theory and missing data","author":"Tsiatis","year":"2007"},{"issue":"1","key":"2026042702520200300_qkag068-B51","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1515\/jci-2013-0022","article-title":"Targeted learning of the mean outcome under an optimal dynamic treatment rule","volume":"3","author":"van der Laan","year":"2015","journal-title":"Journal of Causal Inference"},{"issue":"1","key":"2026042702520200300_qkag068-B52","doi-asserted-by":"publisher","first-page":"Article25","DOI":"10.2202\/1544-6115.1309","article-title":"Super learner","volume":"6","author":"Van der Laan","year":"2007","journal-title":"Statistical Applications in Genetics and Molecular Biology"},{"key":"2026042702520200300_qkag068-B53","author":"van der Vaart","year":"2002"},{"issue":"3","key":"2026042702520200300_qkag068-B54","doi-asserted-by":"publisher","first-page":"399","DOI":"10.1097\/EDE.0000000000000641","article-title":"Outcome-wide epidemiology","volume":"28","author":"VanderWeele","year":"2017","journal-title":"Epidemiology (Cambridge, Mass.)"},{"issue":"8","key":"2026042702520200300_qkag068-B55","doi-asserted-by":"publisher","first-page":"845","DOI":"10.1001\/jamapsychiatry.2016.1243","article-title":"Association between religious service attendance and lower suicide rates among us women","volume":"73","author":"VanderWeele","year":"2016","journal-title":"JAMA Psychiatry"},{"issue":"523","key":"2026042702520200300_qkag068-B56","doi-asserted-by":"publisher","first-page":"1228","DOI":"10.1080\/01621459.2017.1319839","article-title":"Estimation and inference of heterogeneous treatment effects using random forests","volume":"113","author":"Wager","year":"2018","journal-title":"Journal of the American Statistical Association"},{"issue":"15","key":"2026042702520200300_qkag068-B57","doi-asserted-by":"publisher","first-page":"2372","DOI":"10.1093\/bioinformatics\/btx174","article-title":"Mining heterogeneous causal effects for personalized cancer treatment","volume":"33","author":"Zhang","year":"2017","journal-title":"Bioinformatics"},{"issue":"3","key":"2026042702520200300_qkag068-B58","doi-asserted-by":"publisher","first-page":"697","DOI":"10.1111\/j.1541-0420.2011.01712.x","article-title":"Causal inference on quantiles with an obstetric application","volume":"68","author":"Zhang","year":"2012","journal-title":"Biometrics"},{"key":"2026042702520200300_qkag068-B59","author":"Zheng","year":"2010"}],"container-title":["Journal of the Royal Statistical Society Series B: Statistical Methodology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article-pdf\/doi\/10.1093\/jrsssb\/qkag068\/68162976\/qkag068.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article-pdf\/doi\/10.1093\/jrsssb\/qkag068\/68162976\/qkag068.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,27]],"date-time":"2026-04-27T06:52:11Z","timestamp":1777272731000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article\/doi\/10.1093\/jrsssb\/qkag068\/8662882"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,27]]},"references-count":59,"URL":"https:\/\/doi.org\/10.1093\/jrsssb\/qkag068","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,27]]},"article-number":"qkag068"}}