{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T20:22:22Z","timestamp":1790713342769,"version":"4.1.0"},"reference-count":26,"publisher":"The Econometric Society","issue":"1","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["ECTA"],"abstract":"<jats:p>Standard experimental designs are geared toward point estimation and hypothesis testing, while bandit algorithms are geared toward in\u2010sample outcomes. Here, we instead consider treatment assignment in an experiment with several waves for choosing the best among a set of possible policies (treatments) at the end of the experiment. We propose a computationally tractable assignment algorithm that we call \u201cexploration sampling,\u201d where assignment probabilities in each wave are an increasing concave function of the posterior probabilities that each treatment is optimal. We prove an asymptotic optimality result for this algorithm and demonstrate improvements in welfare in calibrated simulations over both non\u2010adaptive designs and bandit algorithms. An application to selecting between six different recruitment strategies for an agricultural extension service in India demonstrates practical feasibility.<\/jats:p>","DOI":"10.3982\/ecta17527","type":"journal-article","created":{"date-parts":[[2021,1,12]],"date-time":"2021-01-12T03:05:05Z","timestamp":1610420705000},"page":"113-132","source":"Crossref","is-referenced-by-count":81,"title":["Adaptive Treatment Assignment in Experiments for Policy Choice"],"prefix":"10.3982","volume":"89","author":[{"given":"Maximilian","family":"Kasy","sequence":"first","affiliation":[{"name":"Department of Economics, Oxford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anja","family":"Sautmann","sequence":"additional","affiliation":[{"name":"Development Research Group, World Bank"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2359","published-online":{"date-parts":[[2021]]},"reference":[{"key":"ecta200244-bib-0001","first-page":"39.1","volume-title":"Conference on Learning Theory","author":"Agrawal","year":"2012"},{"issue":"5","key":"ecta200244-bib-0002","doi-asserted-by":"crossref","first-page":"2383","DOI":"10.1257\/aer.100.5.2383","volume":"100","author":"Ashraf","year":"2010","journal-title":"American Economic Review"},{"key":"ecta200244-bib-0003","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1016\/bs.hefe.2016.10.003","volume-title":"Handbook of Economic Field Experiments","volume":"1","author":"Athey","year":"2017"},{"key":"ecta200244-bib-0004","unstructured":"Bastani, H., M. Bayati, and K. Khosravi (2017): \u201cMostly Exploration-Free Algorithms for Contextual Bandits.\u201d"},{"issue":"1","key":"ecta200244-bib-0005","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1038\/nrd1927","volume":"5","author":"Berry","year":"2006","journal-title":"Nature Reviews Drug Discovery"},{"issue":"5","key":"ecta200244-bib-0006","doi-asserted-by":"crossref","first-page":"1671","DOI":"10.3982\/ECTA10489","volume":"82","author":"Bryan","year":"2014","journal-title":"Econometrica"},{"issue":"1","key":"ecta200244-bib-0007","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1561\/2200000024","volume":"5","author":"Bubeck","year":"2012","journal-title":"Foundations and Trends\u00ae in Machine Learning"},{"issue":"19","key":"ecta200244-bib-0008","doi-asserted-by":"crossref","first-page":"1832","DOI":"10.1016\/j.tcs.2010.12.059","volume":"412","author":"Bubeck","year":"2011","journal-title":"Theoretical Computer Science"},{"issue":"2","key":"ecta200244-bib-0009","doi-asserted-by":"crossref","first-page":"609","DOI":"10.1257\/aer.20130267","volume":"105","author":"Cohen","year":"2015","journal-title":"American Economic Review"},{"key":"ecta200244-bib-0010","unstructured":"Dimakopoulou, M., S. Athey, and G. Imbens (2017): \u201cEstimation Considerations in Contextual Bandits,\u201d. Preprint. arXiv:1711.07077."},{"key":"ecta200244-bib-0011","doi-asserted-by":"crossref","unstructured":"Dimakopoulou, M., Z. Zhou, S. Athey, and G. Imbens (2018): \u201cBalanced Linear Contextual Bandits.\u201d","DOI":"10.1609\/aaai.v33i01.33013445"},{"key":"ecta200244-bib-0012","unstructured":"Dudik, M., D. Hsu, S. Kale, N. Karampatziakis, J. Langford, L. Reyzin, and T. Zhang (2011): \u201cEfficient Optimal Learning for Contextual Bandits,\u201d. Preprint. arXiv:1106.2369."},{"key":"ecta200244-bib-0013","unstructured":"FDA (2018): \u201cAdaptive Designs for Clinical Trials of Drugs and Biologics, Guidance for Industry,\u201d U.S. Department of Health and Human Services, Food and Drug Administration."},{"key":"ecta200244-bib-0014","unstructured":"Frazier, P. I. (2018): \u201cA Tutorial on Bayesian Optimization,\u201d. Preprint. arXiv:1807.02811."},{"key":"ecta200244-bib-0015","doi-asserted-by":"crossref","DOI":"10.1017\/9781139029834","volume-title":"Fundamentals of Nonparametric Bayesian Inference","volume":"44","author":"Ghosal","year":"2017"},{"key":"ecta200244-bib-0016","first-page":"577","volume-title":"Proceedings of the 36th Conference on Winter Simulation","author":"Glynn","year":"2004"},{"key":"ecta200244-bib-0017","doi-asserted-by":"crossref","unstructured":"Kasy, M., and A. Sautmann (2021): \u201cSupplement to \u2018Adaptive Treatment Assignment in Experiments for Policy Choice\u2019,\u201d Econometrica Supplemental Material, 89, https:\/\/doi.org\/10.3982\/ECTA17527.","DOI":"10.3982\/ECTA17527"},{"issue":"2","key":"ecta200244-bib-0018","doi-asserted-by":"crossref","first-page":"591","DOI":"10.3982\/ECTA13288","volume":"86","author":"Kitagawa","year":"2018","journal-title":"Econometrica"},{"key":"ecta200244-bib-0019","unstructured":"Kock, A. B., and M. Thyrsgaard (2017): \u201cOptimal Sequential Treatment Allocation,\u201d. Preprint. arXiv:1705.09952."},{"key":"ecta200244-bib-0020","unstructured":"Kock, A. B., D. Preinerstorfer, and B. Veliyev (2018): \u201cFunctional Sequential Treatment Allocation,\u201d. Preprint. arXiv:1812.09408."},{"issue":"1","key":"ecta200244-bib-0021","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1016\/0196-8858(85)90002-8","volume":"6","author":"Lai","year":"1985","journal-title":"Advances in applied mathematics"},{"key":"ecta200244-bib-0022","first-page":"1417","volume-title":"Conference on Learning Theory","author":"Russo","year":"2016"},{"issue":"1","key":"ecta200244-bib-0023","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1561\/2200000070","volume":"11","author":"Russo","year":"2018","journal-title":"Foundations and Trends\u00ae in Machine Learning"},{"issue":"6","key":"ecta200244-bib-0024","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1002\/asmb.874","volume":"26","author":"Scott","year":"2010","journal-title":"Applied Stochastic Models in Business and Industry"},{"issue":"3\/4","key":"ecta200244-bib-0025","doi-asserted-by":"crossref","first-page":"285","DOI":"10.2307\/2332286","volume":"25","author":"Thompson","year":"1933","journal-title":"Biometrika"},{"issue":"4","key":"ecta200244-bib-0026","doi-asserted-by":"crossref","first-page":"1024","DOI":"10.1214\/aoap\/1177005588","volume":"2","author":"Weber","year":"1992","journal-title":"The Annals of Applied Probability"}],"container-title":["Econometrica"],"original-title":[],"language":"en","deposited":{"date-parts":[[2021,1,12]],"date-time":"2021-01-12T03:16:55Z","timestamp":1610421415000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.econometricsociety.org\/doi\/10.3982\/ECTA17527"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":26,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021]]}},"alternative-id":["10.3982\/ECTA17527"],"URL":"https:\/\/doi.org\/10.3982\/ecta17527","relation":{},"ISSN":["0012-9682"],"issn-type":[{"value":"0012-9682","type":"print"}],"subject":[],"published":{"date-parts":[[2021]]}}}