{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T10:53:52Z","timestamp":1785408832212,"version":"3.56.0"},"reference-count":70,"publisher":"Cambridge University Press (CUP)","license":[{"start":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T00:00:00Z","timestamp":1784764800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["SES-2315600"],"award-info":[{"award-number":["SES-2315600"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Econom. Theory"],"abstract":"<jats:p>We study a class of binary treatment choice problems with partial identification\n                    through the lens of robust (multiple prior) Bayesian analysis. We use a\n                    convenient set of prior distributions to derive ex-ante and ex-post robust Bayes\n                    decision rules, both for decision makers who can randomize and for decision\n                    makers who cannot. Our main messages are as follows: First, ex-ante and ex-post\n                    robust Bayes decision rules do not agree in general, whether or not randomized\n                    rules are allowed. Second, randomized treatment assignment for some data\n                    realizations can be optimal in both ex-ante and, perhaps more surprisingly,\n                    ex-post problems. Therefore, it is usually with loss of generality to exclude\n                    randomized rules from consideration, even when regret is evaluated ex post. We\n                    apply our results to a stylized problem where a policy maker uses experimental\n                    data to choose whether to implement a new policy in a population of interest,\n                    but is concerned about the external validity of the experiment at hand, and to\n                    aggregation of data generated by multiple randomized control trials in different\n                    sites to make a policy choice in a population for which no experimental data are\n                    available.<\/jats:p>","DOI":"10.1017\/s0266466626100486","type":"journal-article","created":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T09:41:19Z","timestamp":1784799679000},"page":"1-37","source":"Crossref","is-referenced-by-count":0,"title":["ROBUST BAYES TREATMENT CHOICE WITH PARTIAL\n                    IDENTIFICATION"],"prefix":"10.1017","author":[{"given":"Andr\u00e9s","family":"Aradillas Fern\u00e1ndez","sequence":"first","affiliation":[{"id":[{"id":"https:\/\/ror.org\/042nb2s44","id-type":"ROR","asserted-by":"publisher"}],"name":"Massachusetts Institute of Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jos\u00e9 Luis","family":"Montiel Olea","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05bnh6r87","id-type":"ROR","asserted-by":"publisher"}],"name":"Cornell University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Qiu","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05bnh6r87","id-type":"ROR","asserted-by":"publisher"}],"name":"Cornell University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J\u00f6rg","family":"Stoye","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05bnh6r87","id-type":"ROR","asserted-by":"publisher"}],"name":"Cornell University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Serdil","family":"Tinda","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/00f54p054","id-type":"ROR","asserted-by":"publisher"}],"name":"Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"56","published-online":{"date-parts":[[2026,7,23]]},"reference":[{"key":"S0266466626100486_r33","unstructured":"Kido, D. (2022). Distributionally robust policy learning with Wasserstein distance. Preprint, arXiv:2205.04637."},{"key":"S0266466626100486_r27","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1016\/bs.hoe.2020.09.001","volume-title":"Handbook of econometrics, volume 7A","volume":"7","author":"Hirano","year":"2020"},{"key":"S0266466626100486_r39","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA13288"},{"key":"S0266466626100486_r3","doi-asserted-by":"crossref","unstructured":"Aradillas Fern\u00e1ndez, A. , Blanchet, J. , Olea, J. L. M. , Qiu, C. , Stoye, J. , & Tan, L. (2025). \u03f5 $\\epsilon$ epsilon -Minimax solutions of statistical decision problems. Preprint, arXiv:2509.08107.","DOI":"10.1145\/3736252.3742598"},{"key":"S0266466626100486_r57","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1951.10500768"},{"key":"S0266466626100486_r16","doi-asserted-by":"publisher","DOI":"10.1524\/strm.1989.7.4.333"},{"key":"S0266466626100486_r32","doi-asserted-by":"publisher","DOI":"10.1007\/s00182-005-0004-3"},{"key":"S0266466626100486_r53","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA16437"},{"key":"S0266466626100486_r58","unstructured":"Schlag, K. H. (2006). ELEVEN\u2014Tests needed for a recommendation. Tech. rep., European University Institute Working Paper, ECO No. 2006\/2. https:\/\/cadmus.eui.eu\/bitstream\/handle\/1814\/3937\/ECO2006-2.pdf"},{"key":"S0266466626100486_r60","doi-asserted-by":"publisher","DOI":"10.1017\/S0266466607070089"},{"key":"S0266466626100486_r24","unstructured":"Guggenberger, P. , Mehta, N. , & Pavlov, N. (2024). Minimax regret treatment rules with finite samples when a quantile is the object of interest. Tech. rep., The Pennsylvania State University."},{"key":"S0266466626100486_r38","unstructured":"Kitagawa, T. , Sakaguchi, S. , & Tetenov, A. (2021). Constrained classification and policy learning. Preprint, https:\/\/arxiv.org\/abs\/2106.12886."},{"key":"S0266466626100486_r8","doi-asserted-by":"publisher","DOI":"10.1080\/03610929208830830"},{"key":"S0266466626100486_r12","volume-title":"The Oxford handbook of Bayesian econometrics","author":"Chamberlain","year":"2011"},{"key":"S0266466626100486_r36","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/asag008"},{"key":"S0266466626100486_r45","doi-asserted-by":"publisher","DOI":"10.1111\/j.1468-0262.2004.00530.x"},{"key":"S0266466626100486_r69","doi-asserted-by":"publisher","DOI":"10.2307\/1969022"},{"key":"S0266466626100486_r18","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177729689"},{"key":"S0266466626100486_r56","doi-asserted-by":"publisher","DOI":"10.1017\/S0266466698144043"},{"key":"S0266466626100486_r22","doi-asserted-by":"publisher","DOI":"10.21033\/wp-2021-11"},{"key":"S0266466626100486_r28","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA21180"},{"key":"S0266466626100486_r43","unstructured":"Lei, L. , Sahoo, R. , & Wager, S. (2023). Policy learning under biased sample selection. Preprint, arXiv:2304.11735."},{"key":"S0266466626100486_r59","doi-asserted-by":"publisher","DOI":"10.1017\/S0266466613000327"},{"key":"S0266466626100486_r50","doi-asserted-by":"publisher","DOI":"10.1017\/S0266466624000197"},{"key":"S0266466626100486_r49","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0000000000001178"},{"key":"S0266466626100486_r55","doi-asserted-by":"crossref","first-page":"755","DOI":"10.3982\/ECTA8360","article-title":"Bayesian and frequentist\n                        inference in partially identified models","volume":"80","author":"Moon","year":"2012","journal-title":"Econometrica"},{"key":"S0266466626100486_r51","doi-asserted-by":"publisher","DOI":"10.1016\/j.jspi.2006.06.037"},{"key":"S0266466626100486_r30","unstructured":"Kallus, N. , & Zhou, A. (2018). Confounding-robust policy improvement. In Advances in neural information processing systems (Vol. 31), 9289\u20139299."},{"key":"S0266466626100486_r63","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2011.06.012"},{"key":"S0266466626100486_r21","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA16773"},{"key":"S0266466626100486_r42","volume-title":"Theory of point estimation","author":"Lehmann","year":"1998"},{"key":"S0266466626100486_r67","doi-asserted-by":"crossref","unstructured":"Vidakovic, B. (2000). \u0393 $\\varGamma$ upper Gamma -minimax: A paradigm for conservative robust Bayesians. In D. R. Insua & F. Ruggeri (Eds.), Robust Bayesian analysis (pp. 241\u2013259). Springer.","DOI":"10.1007\/978-1-4612-1306-2_13"},{"key":"S0266466626100486_r11","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1970.10481080"},{"key":"S0266466626100486_r70","unstructured":"Yata, K. (2021). Optimal decision rules under partial identification. Preprint, arXiv:2111.04926."},{"key":"S0266466626100486_r13","article-title":"A note on minimax regret\n                        rules with multiple treatments in finite samples","author":"Chen","year":"2025","journal-title":"Econometric Theory"},{"key":"S0266466626100486_r35","doi-asserted-by":"crossref","unstructured":"Kitagawa, T. (2012). Estimation and inference for set-identified parameters using posterior lower probability. Manuscript, UCL.","DOI":"10.1920\/wp.cem.2011.1611"},{"key":"S0266466626100486_r31","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177728259"},{"key":"S0266466626100486_r46","volume-title":"Social choice with partial knowledge\n                        of treatment response","author":"Manski","year":"2005"},{"key":"S0266466626100486_r48","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2006.06.006"},{"key":"S0266466626100486_r1","unstructured":"Adjaho, C. , & Christensen, T. (2022). Externally valid treatment choice. Preprint, arXiv:2205.05561."},{"key":"S0266466626100486_r68","doi-asserted-by":"publisher","DOI":"10.1016\/j.jet.2005.10.008"},{"key":"S0266466626100486_r54","doi-asserted-by":"publisher","DOI":"10.1093\/restud\/rdag015"},{"key":"S0266466626100486_r25","doi-asserted-by":"publisher","DOI":"10.1016\/j.jet.2007.07.001"},{"key":"S0266466626100486_r15","unstructured":"D\u2019Adamo, R. (2021). Policy learning under ambiguity. Preprint, arXiv:2111.10904."},{"key":"S0266466626100486_r40","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.2019.1688664"},{"key":"S0266466626100486_r65","unstructured":"Tetenov, A. (2012a). Measuring precision of statistical inference on partially identified parameters. Discussion Paper, Collegio Carlo Alberto, Torino."},{"key":"S0266466626100486_r4","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA15732"},{"key":"S0266466626100486_r34","unstructured":"Kido, D. (2023). Locally asymptotically minimax statistical treatment rules under partial identification. Preprint, arXiv:2311.08958."},{"key":"S0266466626100486_r9","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2011.11.007"},{"key":"S0266466626100486_r14","doi-asserted-by":"crossref","unstructured":"Christensen, T. , Moon, H. R. , & Schorfheide, F. (2026). Optimal discrete decisions when payoffs are partially identified. Review of Economic Studies. Preprint, arXiv:2204.11748.","DOI":"10.1093\/restud\/rdag017"},{"key":"S0266466626100486_r23","unstructured":"Guggenberger, P. , & Huang, J. (2025). On the numerical approximation of minimax regret rules via fictitious play. Preprint, arXiv:2503.10932."},{"key":"S0266466626100486_r41","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2021.09.009"},{"key":"S0266466626100486_r61","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2009.02.013"},{"key":"S0266466626100486_r10","doi-asserted-by":"publisher","DOI":"10.1111\/j.1468-0297.2006.01130.x"},{"key":"S0266466626100486_r5","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2025.2489135"},{"key":"S0266466626100486_r64","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-economics-080511-110959"},{"key":"S0266466626100486_r26","doi-asserted-by":"crossref","first-page":"1683","DOI":"10.3982\/ECTA6630","article-title":"Asymptotics for statistical\n                        treatment rules","volume":"77","author":"Hirano","year":"2009","journal-title":"Econometrica"},{"key":"S0266466626100486_r66","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2011.06.013"},{"key":"S0266466626100486_r2","doi-asserted-by":"publisher","DOI":"10.1007\/s40505-018-0147-6"},{"key":"S0266466626100486_r29","unstructured":"Ishihara, T. , & Kitagawa, T. (2021). Evidence aggregation for treatment choice. Preprint, arXiv:2108.06473."},{"key":"S0266466626100486_r62","doi-asserted-by":"publisher","DOI":"10.1016\/j.jet.2011.10.004"},{"key":"S0266466626100486_r6","unstructured":"Ben-Michael, E. , Imai, K. , & Jiang, Z. (2022). Policy learning with asymmetric utilities. Preprint, arXiv:2206.10479."},{"key":"S0266466626100486_r17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2004.04.006"},{"key":"S0266466626100486_r37","doi-asserted-by":"publisher","DOI":"10.1007\/s42973-023-00144-3"},{"key":"S0266466626100486_r52","doi-asserted-by":"publisher","DOI":"10.1007\/s42973-023-00145-2"},{"key":"S0266466626100486_r7","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4757-4286-2"},{"key":"S0266466626100486_r20","volume-title":"Mathematical statistics: A decision\n                        theoretic approach","volume":"7","author":"Ferguson","year":"1967"},{"key":"S0266466626100486_r44","doi-asserted-by":"publisher","DOI":"10.1016\/S0304-4076(99)00045-7"},{"key":"S0266466626100486_r47","volume-title":"Identification for prediction and\n                        decision","author":"Manski","year":"2007a"},{"key":"S0266466626100486_r19","doi-asserted-by":"publisher","DOI":"10.1016\/S0022-0531(03)00097-8"}],"container-title":["Econometric Theory"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S0266466626100486","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T10:24:16Z","timestamp":1785147856000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S0266466626100486\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,23]]},"references-count":70,"alternative-id":["S0266466626100486"],"URL":"https:\/\/doi.org\/10.1017\/s0266466626100486","relation":{},"ISSN":["0266-4666","1469-4360"],"issn-type":[{"value":"0266-4666","type":"print"},{"value":"1469-4360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,23]]}}}