{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T16:36:10Z","timestamp":1791909370812,"version":"4.3.4"},"reference-count":91,"publisher":"The Econometric Society","issue":"1","funder":[{"DOI":"10.13039\/100000006","name":"Office of Naval Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["ECTA"],"abstract":"<jats:p>In many areas, practitioners seek to use observational data to learn a treatment assignment policy that satisfies application\u2010specific constraints, such as budget, fairness, simplicity, or other functional form constraints. For example, policies may be restricted to take the form of decision trees based on a limited set of easily observable individual characteristics. We propose a new approach to this problem motivated by the theory of semiparametrically efficient estimation. Our method can be used to optimize either binary treatments or infinitesimal nudges to continuous treatments, and can leverage observational data where causal effects are identified using a variety of strategies, including selection on observables and instrumental variables. Given a doubly robust estimator of the causal effect of assigning everyone to treatment, we develop an algorithm for choosing whom to treat, and establish strong guarantees for the asymptotic utilitarian regret of the resulting policy.<\/jats:p>","DOI":"10.3982\/ecta15732","type":"journal-article","created":{"date-parts":[[2021,1,12]],"date-time":"2021-01-12T03:05:05Z","timestamp":1610420705000},"page":"133-161","source":"Crossref","is-referenced-by-count":341,"title":["Policy Learning With Observational Data"],"prefix":"10.3982","volume":"89","author":[{"given":"Susan","family":"Athey","sequence":"first","affiliation":[{"name":"Stanford Graduate School of Business, Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefan","family":"Wager","sequence":"additional","affiliation":[{"name":"Stanford Graduate School of Business, Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2359","published-online":{"date-parts":[[2021]]},"reference":[{"issue":"2","key":"ecta200236-bib-0001","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1016\/S0304-4076(02)00201-4","volume":"113","author":"Abadie","year":"2003","journal-title":"Journal of Econometrics"},{"issue":"1","key":"ecta200236-bib-0002","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1016\/j.jeconom.2007.01.013","volume":"141","author":"Ai","year":"2007","journal-title":"Journal of Econometrics"},{"key":"ecta200236-bib-0003","unstructured":"Armstrong, T. B., and M. Koles\u00e1r (2017): \u201cFinite-Sample Optimal Estimation and Inference on Average Treatment Effects Under Unconfoundedness,\u201d arXiv preprint arXiv:1712.04594."},{"key":"ecta200236-bib-0004","doi-asserted-by":"crossref","unstructured":"Armstrong, T. B., and S. Shen (2013): \u201cInference on Optimal Treatment Assignments,\u201d Technical report, Cowles Foundation for Research in Economics, Yale University.","DOI":"10.2139\/ssrn.2355423"},{"issue":"4","key":"ecta200236-bib-0005","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1093\/pan\/mpt013","volume":"21","author":"Aronow","year":"2013","journal-title":"Political Analysis"},{"issue":"27","key":"ecta200236-bib-0006","doi-asserted-by":"crossref","first-page":"7353","DOI":"10.1073\/pnas.1510489113","volume":"113","author":"Athey","year":"2016","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"ecta200236-bib-0007","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1353\/obs.2019.0001","volume":"5","author":"Athey","year":"2019","journal-title":"Observational Studies"},{"key":"ecta200236-bib-0008","doi-asserted-by":"crossref","unstructured":"Athey, S., and S. Wager (2021): \u201cSupplement to \u2018Policy Learning With Observational Data\u2019,\u201d Econometrica Supplemental Material, 89, https:\/\/doi.org\/10.3982\/ECTA15732.","DOI":"10.3982\/ECTA15732"},{"issue":"2","key":"ecta200236-bib-0009","doi-asserted-by":"crossref","first-page":"1148","DOI":"10.1214\/18-AOS1709","volume":"47","author":"Athey","year":"2019","journal-title":"The Annals of Statistics"},{"issue":"1","key":"ecta200236-bib-0010","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1287\/opre.2018.1757","volume":"67","author":"Ban","year":"2019","journal-title":"Operations Research"},{"key":"ecta200236-bib-0011","first-page":"463","volume":"3","author":"Bartlett","year":"2002","journal-title":"Journal of Machine Learning Research"},{"issue":"3","key":"ecta200236-bib-0012","doi-asserted-by":"crossref","first-page":"311","DOI":"10.1007\/s00440-005-0462-3","volume":"135","author":"Bartlett","year":"2006","journal-title":"Probability Theory and Related Fields"},{"issue":"4","key":"ecta200236-bib-0013","doi-asserted-by":"crossref","first-page":"1497","DOI":"10.1214\/009053605000000282","volume":"33","author":"Bartlett","year":"2005","journal-title":"Annals of Statistics"},{"issue":"7","key":"ecta200236-bib-0014","doi-asserted-by":"crossref","first-page":"1039","DOI":"10.1007\/s10994-017-5633-9","volume":"106","author":"Bertsimas","year":"2017","journal-title":"Machine Learning"},{"issue":"3","key":"ecta200236-bib-0015","doi-asserted-by":"crossref","first-page":"1025","DOI":"10.1287\/mnsc.2018.3253","volume":"66","author":"Bertsimas","year":"2020","journal-title":"Management Science"},{"key":"ecta200236-bib-0016","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1145\/1557019.1557040","volume-title":"Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Beygelzimer","year":"2009"},{"issue":"1","key":"ecta200236-bib-0017","doi-asserted-by":"crossref","first-page":"168","DOI":"10.1016\/j.jeconom.2011.11.007","volume":"167","author":"Bhattacharya","year":"2012","journal-title":"Journal of Econometrics"},{"key":"ecta200236-bib-0018","volume-title":"Efficient and Adaptive Estimation for Semiparametric Models","author":"Bickel","year":"1998"},{"issue":"1","key":"ecta200236-bib-0019","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","volume":"45","author":"Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"ecta200236-bib-0020","volume-title":"Classification and Regression Trees","author":"Breiman","year":"1984"},{"issue":"3","key":"ecta200236-bib-0021","doi-asserted-by":"crossref","first-page":"331","DOI":"10.1007\/s10208-006-0196-8","volume":"7","author":"Caponnetto","year":"2007","journal-title":"Foundations of Computational Mathematics"},{"key":"ecta200236-bib-0022","volume-title":"The Oxford Handbook of Bayesian Econometrics","author":"Chamberlain","year":"2011"},{"issue":"1","key":"ecta200236-bib-0023","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/j.jeconom.2018.05.001","volume":"206","author":"Chen","year":"2018","journal-title":"Journal of Econometrics"},{"key":"ecta200236-bib-0024","doi-asserted-by":"crossref","first-page":"5549","DOI":"10.1016\/S1573-4412(07)06076-X","volume":"6","author":"Chen","year":"2007","journal-title":"Handbook of Econometrics"},{"issue":"2","key":"ecta200236-bib-0025","doi-asserted-by":"crossref","first-page":"808","DOI":"10.1214\/009053607000000947","volume":"36","author":"Chen","year":"2008","journal-title":"The Annals of Statistics"},{"issue":"1","key":"ecta200236-bib-0026","doi-asserted-by":"crossref","first-page":"C1","DOI":"10.1111\/ectj.12097","volume":"21","author":"Chernozhukov","year":"2018","journal-title":"The Econometrics Journal"},{"key":"ecta200236-bib-0027","unstructured":"Chernozhukov, V., J. C. Escanciano, H. Ichimura, W. K. Newey, and J. M. Robins (2016): \u201cLocally Robust Semiparametric Estimation,\u201d arXiv preprint arXiv:1608.00033."},{"issue":"3","key":"ecta200236-bib-0028","first-page":"273","volume":"20","author":"Cortes","year":"1995","journal-title":"Machine learning"},{"key":"ecta200236-bib-0029","first-page":"442","volume-title":"Advances in Neural Information Processing Systems","author":"Cortes","year":"2010"},{"issue":"1","key":"ecta200236-bib-0030","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1016\/j.jeconom.2004.04.006","volume":"125","author":"Dehejia","year":"2005","journal-title":"Journal of Econometrics"},{"issue":"7","key":"ecta200236-bib-0031","doi-asserted-by":"crossref","first-page":"1011","DOI":"10.1016\/0031-3203(94)00141-8","volume":"28","author":"Devroye","year":"1995","journal-title":"Pattern recognition"},{"key":"ecta200236-bib-0032","first-page":"1097","volume-title":"Proceedings of the 28th International Conference on Machine Learning","author":"Dud\u00edk","year":"2011"},{"issue":"1","key":"ecta200236-bib-0033","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jeconom.2015.06.017","volume":"189","author":"Farrell","year":"2015","journal-title":"Journal of Econometrics"},{"key":"ecta200236-bib-0034","doi-asserted-by":"crossref","unstructured":"Farrell, M. H., T. Liang, and S. Misra (2020): \u201cDeep Neural Networks for Estimation and Inference,\u201d Econometrica. (forthcoming).","DOI":"10.3982\/ECTA16901"},{"key":"ecta200236-bib-0035","unstructured":"Fithian, W., D. Sun, and J. Taylor (2014): \u201cOptimal Inference After Model Selection,\u201d arXiv preprint arXiv:1410.2597."},{"issue":"3","key":"ecta200236-bib-0036","doi-asserted-by":"crossref","first-page":"1143","DOI":"10.1214\/009117906000000070","volume":"34","author":"Gin\u00e9","year":"2006","journal-title":"The Annals of Probability"},{"issue":"2","key":"ecta200236-bib-0037","doi-asserted-by":"crossref","first-page":"315","DOI":"10.2307\/2998560","volume":"66","author":"Hahn","year":"1998","journal-title":"Econometrica"},{"issue":"5","key":"ecta200236-bib-0038","doi-asserted-by":"crossref","first-page":"1683","DOI":"10.3982\/ECTA6630","volume":"77","author":"Hirano","year":"2009","journal-title":"Econometrica"},{"issue":"4","key":"ecta200236-bib-0039","doi-asserted-by":"crossref","first-page":"1161","DOI":"10.1111\/1468-0262.00442","volume":"71","author":"Hirano","year":"2003","journal-title":"Econometrica"},{"key":"ecta200236-bib-0040","unstructured":"Hirshberg, D. A., and S. Wager (2018): \u201cAugmented Minimax Linear Estimation,\u201d arXiv preprint arXiv:1712.00038."},{"issue":"3","key":"ecta200236-bib-0041","doi-asserted-by":"crossref","first-page":"521","DOI":"10.1086\/505050","volume":"24","author":"Hotz","year":"2006","journal-title":"Journal of Labor Economics"},{"issue":"482","key":"ecta200236-bib-0042","doi-asserted-by":"crossref","first-page":"832","DOI":"10.1198\/016214508000000292","volume":"103","author":"Hudgens","year":"2008","journal-title":"Journal of the American Statistical Association"},{"issue":"2","key":"ecta200236-bib-0043","doi-asserted-by":"crossref","first-page":"467","DOI":"10.2307\/2951620","volume":"62","author":"Imbens","year":"1994","journal-title":"Econometrica"},{"key":"ecta200236-bib-0044","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9781139025751","volume-title":"Causal Inference in Statistics, Social, and Biomedical Sciences","author":"Imbens","year":"2015"},{"key":"ecta200236-bib-0045","first-page":"8895","volume-title":"Advances in Neural Information Processing Systems","author":"Kallus","year":"2018"},{"key":"ecta200236-bib-0046","unstructured":"Kallus, N., and A. Zhou (2020): \u201cConfounding-Robust Policy Improvement,\u201d Management Science. (forthcoming)."},{"issue":"1","key":"ecta200236-bib-0047","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1162\/REST_a_00528","volume":"98","author":"Kasy","year":"2016","journal-title":"Review of Economics and Statistics"},{"issue":"2","key":"ecta200236-bib-0048","doi-asserted-by":"crossref","first-page":"591","DOI":"10.3982\/ECTA13288","volume":"86","author":"Kitagawa","year":"2018","journal-title":"Econometrica"},{"issue":"3","key":"ecta200236-bib-0049","doi-asserted-by":"crossref","first-page":"2162","DOI":"10.1214\/19-AIHP1034","volume":"56","author":"Luedtke","year":"2020","journal-title":"Annales de l'Institut Henri Poincar\u00e9, Probabilit\u00e9s et Statistiques"},{"issue":"2","key":"ecta200236-bib-0050","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1214\/15-AOS1384","volume":"44","author":"Luedtke","year":"2016","journal-title":"The Annals of Statistics"},{"issue":"4","key":"ecta200236-bib-0051","doi-asserted-by":"crossref","first-page":"1221","DOI":"10.1111\/j.1468-0262.2004.00530.x","volume":"72","author":"Manski","year":"2004","journal-title":"Econometrica"},{"key":"ecta200236-bib-0052","doi-asserted-by":"crossref","DOI":"10.2307\/j.ctv219kxm0","volume-title":"Identification for Prediction and Decision","author":"Manski","year":"2009"},{"issue":"1","key":"ecta200236-bib-0053","doi-asserted-by":"crossref","first-page":"S1","DOI":"10.1111\/j.1368-423X.2012.00368.x","volume":"16","author":"Manski","year":"2013","journal-title":"The Econometrics Journal"},{"issue":"4","key":"ecta200236-bib-0054","doi-asserted-by":"crossref","first-page":"997","DOI":"10.1111\/1468-0262.00144","volume":"68","author":"Manski","year":"2000","journal-title":"Econometrica"},{"key":"ecta200236-bib-0055","volume-title":"Conference on Learning Theory","author":"Maurer","year":"2009"},{"key":"ecta200236-bib-0056","unstructured":"Mbakop, E., and M. Tabord-Meehan (2016): \u201cModel Selection for Treatment Choice: Penalized Welfare Maximization,\u201d arXiv preprint arXiv:1609.03167."},{"issue":"6","key":"ecta200236-bib-0057","doi-asserted-by":"crossref","first-page":"1349","DOI":"10.2307\/2951752","volume":"62","author":"Newey","year":"1994","journal-title":"Econometrica"},{"key":"ecta200236-bib-0058","first-page":"1","volume":"10","author":"Neyman","year":"1923","journal-title":"Roczniki Nauk Rolniczych"},{"issue":"2","key":"ecta200236-bib-0059","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1007\/s10959-007-0058-1","volume":"20","author":"Nickl","year":"2007","journal-title":"Journal of Theoretical Probability"},{"key":"ecta200236-bib-0060","doi-asserted-by":"crossref","unstructured":"Nie, X., and S. Wager (2020): \u201cQuasi-Oracle Estimation of Heterogeneous Treatment Effects,\u201d Biometrika. (forthcoming).","DOI":"10.1093\/biomet\/asaa076"},{"key":"ecta200236-bib-0061","doi-asserted-by":"crossref","unstructured":"Nie, X., E. Brunskill, and S. Wager (2019): \u201cLearning When-to-Treat Policies,\u201d Journal of the American Statistical Association. (forthcoming).","DOI":"10.1080\/01621459.2020.1831925"},{"issue":"6","key":"ecta200236-bib-0062","doi-asserted-by":"crossref","first-page":"1403","DOI":"10.2307\/1913713","volume":"57","author":"Powell","year":"1989","journal-title":"Econometrica"},{"issue":"2","key":"ecta200236-bib-0063","doi-asserted-by":"crossref","first-page":"1180","DOI":"10.1214\/10-AOS864","volume":"39","author":"Qian","year":"2011","journal-title":"Annals of Statistics"},{"key":"ecta200236-bib-0064","volume-title":"R: A Language and Environment for Statistical Computing","year":"2019"},{"key":"ecta200236-bib-0065","unstructured":"Rai, Y. (2018): \u201cStatistical Inference for Treatment Assignment Policies,\u201d Unpublished Manuscript."},{"issue":"1","key":"ecta200236-bib-0066","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1080\/01621459.1995.10476494","volume":"90","author":"Robins","year":"1995","journal-title":"Journal of the American Statistical Association"},{"issue":"427","key":"ecta200236-bib-0067","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1080\/01621459.1994.10476818","volume":"89","author":"Robins","year":"1994","journal-title":"Journal of the American Statistical Association"},{"issue":"5","key":"ecta200236-bib-0068","doi-asserted-by":"crossref","first-page":"1951","DOI":"10.1214\/16-AOS1515","volume":"45","author":"Robins","year":"2017","journal-title":"The Annals of Statistics"},{"issue":"4","key":"ecta200236-bib-0069","doi-asserted-by":"crossref","first-page":"931","DOI":"10.2307\/1912705","volume":"56","author":"Robinson","year":"1988","journal-title":"Econometrica"},{"key":"ecta200236-bib-0070","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-3692-2","volume-title":"Observational Studies","author":"Rosenbaum","year":"2002"},{"key":"ecta200236-bib-0071","unstructured":"Rothe, C. (2018): \u201cFlexible Covariate Adjustments in Randomized Experiments\u201d, https:\/\/madoc.bib.uni-mannheim.de\/52249\/."},{"issue":"5","key":"ecta200236-bib-0072","doi-asserted-by":"crossref","first-page":"688","DOI":"10.1037\/h0037350","volume":"66","author":"Rubin","year":"1974","journal-title":"Journal of Educational Psychology"},{"issue":"3","key":"ecta200236-bib-0073","doi-asserted-by":"crossref","first-page":"1139","DOI":"10.1214\/aos\/1176350055","volume":"14","author":"Schick","year":"1986","journal-title":"The Annals of Statistics"},{"issue":"1","key":"ecta200236-bib-0074","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.jeconom.2009.02.013","volume":"151","author":"Stoye","year":"2009","journal-title":"Journal of Econometrics"},{"issue":"1","key":"ecta200236-bib-0075","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1016\/j.jeconom.2011.06.012","volume":"166","author":"Stoye","year":"2012","journal-title":"Journal of Econometrics"},{"issue":"50","key":"ecta200236-bib-0076","doi-asserted-by":"crossref","DOI":"10.21105\/joss.02232","volume":"5","author":"Sverdrup","year":"2020","journal-title":"Journal of Open Source Software"},{"key":"ecta200236-bib-0077","first-page":"1731","volume":"16","author":"Swaminathan","year":"2015","journal-title":"Journal of Machine Learning Research"},{"issue":"1","key":"ecta200236-bib-0078","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1016\/j.jeconom.2011.06.013","volume":"166","author":"Tetenov","year":"2012","journal-title":"Journal of Econometrics"},{"key":"ecta200236-bib-0079","first-page":"2139","volume-title":"International Conference on Machine Learning","author":"Thomas","year":"2016"},{"key":"ecta200236-bib-0080","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4419-9782-1","volume-title":"Targeted Learning: Causal Inference for Observational and Experimental Data","author":"Laan","year":"2011"},{"issue":"3","key":"ecta200236-bib-0081","doi-asserted-by":"crossref","first-page":"373","DOI":"10.1524\/stnd.2006.24.3.373","volume":"24","author":"Laan","year":"2006","journal-title":"Statistics & Decisions"},{"issue":"1","key":"ecta200236-bib-0082","volume":"6","author":"Laan","year":"2007","journal-title":"Statistical applications in genetics and molecular biology"},{"key":"ecta200236-bib-0083","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-3264-1","volume-title":"The Nature of Statistical Learning Theory","author":"Vapnik","year":"2000"},{"issue":"4","key":"ecta200236-bib-0084","first-page":"1678","volume":"29","author":"Wager","year":"2019","journal-title":"Statistica Sinica"},{"issue":"45","key":"ecta200236-bib-0085","doi-asserted-by":"crossref","first-page":"12673","DOI":"10.1073\/pnas.1614732113","volume":"113","author":"Wager","year":"2016","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"ecta200236-bib-0086","doi-asserted-by":"crossref","DOI":"10.1017\/9781108627771","volume-title":"High-Dimensional Statistics: A Non-Asymptotic Viewpoint","volume":"48","author":"Wainwright","year":"2019"},{"issue":"1","key":"ecta200236-bib-0087","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1002\/sta.411","volume":"1","author":"Zhang","year":"2012","journal-title":"Stat"},{"issue":"3","key":"ecta200236-bib-0088","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1093\/biomet\/ast014","volume":"100","author":"Zhang","year":"2013","journal-title":"Biometrika"},{"issue":"499","key":"ecta200236-bib-0089","doi-asserted-by":"crossref","first-page":"1106","DOI":"10.1080\/01621459.2012.695674","volume":"107","author":"Zhao","year":"2012","journal-title":"Journal of the American Statistical Association"},{"key":"ecta200236-bib-0090","doi-asserted-by":"crossref","first-page":"459","DOI":"10.1007\/978-1-4419-9782-1_27","volume-title":"Targeted Learning","author":"Zheng","year":"2011"},{"key":"ecta200236-bib-0091","unstructured":"Zhou, Z., S. Athey, and S. Wager (2018): \u201cOffline Multi-Action Policy Learning: Generalization and Optimization,\u201d arXiv preprint arXiv:1810.04778."}],"container-title":["Econometrica"],"original-title":[],"language":"en","deposited":{"date-parts":[[2022,12,11]],"date-time":"2022-12-11T03:29:07Z","timestamp":1670729347000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.econometricsociety.org\/doi\/10.3982\/ECTA15732"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":91,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021]]}},"alternative-id":["10.3982\/ECTA15732"],"URL":"https:\/\/doi.org\/10.3982\/ecta15732","relation":{},"ISSN":["0012-9682"],"issn-type":[{"value":"0012-9682","type":"print"}],"subject":[],"published":{"date-parts":[[2021]]}}}