{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T21:19:43Z","timestamp":1790025583619,"version":"4.0.1"},"reference-count":64,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2023,3,3]],"date-time":"2023-03-03T00:00:00Z","timestamp":1677801600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,11,15]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Causal effects are often characterized with averages, which can give an incomplete picture of the underlying counterfactual distributions. Here we consider estimating the entire counterfactual density and generic functionals thereof. We focus on two kinds of target parameters: density approximations and the distance between counterfactual densities. We study nonparametric efficiency bounds, giving results for smooth but otherwise generic models and distances. Importantly, we show how these bounds connect to means of particular nontrivial functions of counterfactuals, linking the problems of density and mean estimation. We propose doubly robust-style estimators, and study their rates of convergence, showing that they can be optimally efficient in large nonparametric models. We also give analogous methods for model selection and aggregation, when many models may be available and of interest. Our results all hold for generic models and distances, but we highlight results for L2 projections on linear models and Kullbach\u2013Leibler projections on exponential families. Finally, we illustrate our method by estimating the density of the CD4 count among patients with HIV, had all been treated with combination therapy versus zidovudine alone, as well as a density effect. Our methods are implemented in the R package npcausal on GitHub.<\/jats:p>","DOI":"10.1093\/biomet\/asad017","type":"journal-article","created":{"date-parts":[[2023,3,4]],"date-time":"2023-03-04T10:30:18Z","timestamp":1677925818000},"page":"875-896","source":"Crossref","is-referenced-by-count":12,"title":["Semiparametric counterfactual density estimation"],"prefix":"10.1093","volume":"110","author":[{"given":"E H","family":"Kennedy","sequence":"first","affiliation":[{"name":"Department of Statistics and Data Science, Carnegie Mellon University , 5000 Forbes Avenue , Pittsburgh, Pennsylvania 15213 U.S.A"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S","family":"Balakrishnan","sequence":"additional","affiliation":[{"name":"Department of Statistics and Data Science, Carnegie Mellon University , 5000 Forbes Avenue , Pittsburgh, Pennsylvania 15213 U.S.A"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"L A","family":"Wasserman","sequence":"additional","affiliation":[{"name":"Department of Statistics and Data Science, Carnegie Mellon University , 5000 Forbes Avenue , Pittsburgh, Pennsylvania 15213 U.S.A"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2023,3,3]]},"reference":[{"key":"2023111615233105100_asad017-B1","doi-asserted-by":"crossref","first-page":"284","DOI":"10.1198\/016214502753479419","article-title":"Bootstrap tests for distributional treatment effects in instrumental variable models","volume":"97","author":"Abadie","year":"2002","journal-title":"J. Am. Statist. Assoc"},{"key":"2023111615233105100_asad017-B2","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1111\/j.2517-6161.1966.tb00626.x","article-title":"A general class of coefficients of divergence of one distribution from another","volume":"28","author":"Ali","year":"1966","journal-title":"J. R. Statist. Soc. B"},{"key":"2023111615233105100_asad017-B3","doi-asserted-by":"crossref","first-page":"445","DOI":"10.1214\/aos\/1176343842","article-title":"Minimum Hellinger distance estimates for parametric models","volume":"5","author":"Beran","year":"1977","journal-title":"Ann. Statist."},{"key":"2023111615233105100_asad017-B4","volume-title":"Efficient and Adaptive Estimation for Semiparametric Models","author":"Bickel","year":"1993"},{"key":"2023111615233105100_asad017-B5","first-page":"381","article-title":"Estimating integrated squared density derivatives: sharp best order of convergence estimates","volume":"4","author":"Bickel","year":"1988","journal-title":"Sankhy\u0101"},{"key":"2023111615233105100_asad017-B6","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1214\/18-STS693","article-title":"Models as approximations I: consequences illustrated with linear regression","volume":"34","author":"Buja","year":"2019","journal-title":"Statist. Sci."},{"key":"2023111615233105100_asad017-B7","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1214\/18-STS694","article-title":"Models as approximations II: a model-free theory of parametric regression","volume":"34","author":"Buja","year":"2019","journal-title":"Statist. Sci."},{"key":"2023111615233105100_asad017-B8","doi-asserted-by":"crossref","first-page":"347","DOI":"10.1093\/biostatistics\/3.3.347","article-title":"A Monte Carlo EM algorithm for generalized linear mixed models with flexible random effects distribution","volume":"3","author":"Chen","year":"2002","journal-title":"Biostatistics"},{"key":"2023111615233105100_asad017-B9","doi-asserted-by":"crossref","first-page":"C1","DOI":"10.1111\/ectj.12097","article-title":"Double\/debiased machine learning for treatment and structural parameters","volume":"21","author":"Chernozhukov","year":"2018","journal-title":"Economet. J"},{"key":"2023111615233105100_asad017-B10","author":"Chernozhukov","year":"2022"},{"key":"2023111615233105100_asad017-B11","doi-asserted-by":"crossref","first-page":"2205","DOI":"10.3982\/ECTA10582","article-title":"Inference on counterfactual distributions","volume":"81","author":"Chernozhukov","year":"2013","journal-title":"Econometrica"},{"key":"2023111615233105100_asad017-B12","doi-asserted-by":"crossref","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"},{"key":"2023111615233105100_asad017-B13","first-page":"229","article-title":"Information-type measures of difference of probability distributions and indirect observation","volume":"2","author":"Csisz\u00e1r","year":"1967","journal-title":"Studia Sci. Math. Hungarica"},{"key":"2023111615233105100_asad017-B14","doi-asserted-by":"crossref","first-page":"1793","DOI":"10.1111\/rssa.12548","article-title":"A non-parametric projection-based estimator for the probability of causation, with application to water sanitation in Kenya","volume":"183","author":"Cuellar","year":"2020","journal-title":"J. R. Statist. Soc. A"},{"key":"2023111615233105100_asad017-B15","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/j.jspi.2017.05.001","article-title":"Efficient estimation of quantiles in missing data models","volume":"190","author":"D\u00edaz","year":"2017","journal-title":"J. Statist. Plan. Infer"},{"key":"2023111615233105100_asad017-B16","doi-asserted-by":"publisher","DOI":"10.2202\/1557-4679.1356","article-title":"Super learner based conditional density estimation with application to marginal structural models","volume":"7","author":"D\u00edaz","year":"2011","journal-title":"Int. J. Biostatistics"},{"key":"2023111615233105100_asad017-B17","doi-asserted-by":"crossref","first-page":"1001","DOI":"10.2307\/2171954","article-title":"Labor market institutions and the distribution of wages, 1973-1992: a semiparametric approach","volume":"64","author":"DiNardo","year":"1996","journal-title":"Econometrica"},{"key":"2023111615233105100_asad017-B18","volume-title":"Large-Scale Inference: Empirical Bayes Methods for Estimation, Testing, and Prediction","author":"Efron","year":"2012"},{"key":"2023111615233105100_asad017-B19","doi-asserted-by":"crossref","first-page":"2431","DOI":"10.1214\/aos\/1032181161","article-title":"Using specially designed exponential families for density estimation","volume":"24","author":"Efron","year":"1996","journal-title":"Ann. Statist."},{"key":"2023111615233105100_asad017-B20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jeconom.2015.06.017","article-title":"Robust inference on average treatment effects with possibly more covariates than observations","volume":"189","author":"Farrell","year":"2015","journal-title":"J. Economet"},{"key":"2023111615233105100_asad017-B21","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1111\/j.1468-0262.2007.00738.x","article-title":"Efficient semiparametric estimation of quantile treatment effects","volume":"75","author":"Firpo","year":"2007","journal-title":"Econometrica"},{"key":"2023111615233105100_asad017-B22","first-page":"1","volume-title":"Handbook of Labor Economics","author":"Fortin","year":"2011"},{"key":"2023111615233105100_asad017-B23","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1080\/07350015.2013.803869","article-title":"Unconditional quantile treatment effects under endogeneity","volume":"31","author":"Fr\u00f6lich","year":"2013","journal-title":"J. Bus. Econ. Statist"},{"key":"2023111615233105100_asad017-B24","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.1056\/NEJM199610103351501","article-title":"A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millimeter","volume":"335","author":"Hammer","year":"1996","journal-title":"N. Engl. J. Med"},{"key":"2023111615233105100_asad017-B25","article-title":"Nonparametric conditional density estimation","author":"Hansen","year":"2004","journal-title":"Neural Computation 21"},{"key":"2023111615233105100_asad017-B26","first-page":"221","volume-title":"Proc. 5th Berkeley Symp. Math. Statist. Prob","author":"Huber","year":"1967"},{"key":"2023111615233105100_asad017-B27","doi-asserted-by":"crossref","first-page":"2800","DOI":"10.1214\/17-EJS1302","article-title":"Converting high-dimensional regression to high-dimensional conditional density estimation","volume":"11","author":"Izbicki","year":"2017","journal-title":"Electron. J. Statist."},{"key":"2023111615233105100_asad017-B28","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1111\/rssb.12300","article-title":"Robust causal inference with continuous instruments using the local instrumental variable curve","volume":"81","author":"Kennedy","year":"2019","journal-title":"J. R. Statist. Soc. B"},{"key":"2023111615233105100_asad017-B29","author":"Kim","year":"2021"},{"key":"2023111615233105100_asad017-B30","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1111\/rssb.12299","article-title":"An omnibus non-parametric test of equality in distribution for unknown functions","volume":"81","author":"Luedtke","year":"2019","journal-title":"J. R. Statist. Soc. B"},{"key":"2023111615233105100_asad017-B31","doi-asserted-by":"crossref","first-page":"445","DOI":"10.1002\/jae.788","article-title":"Counterfactual decomposition of changes in wage distributions using quantile regression","volume":"20","author":"Machado","year":"2005","journal-title":"J. Appl. Economet"},{"key":"2023111615233105100_asad017-B32","doi-asserted-by":"crossref","first-page":"577","DOI":"10.1016\/j.labeco.2005.05.006","article-title":"Decomposition of differences in distribution using quantile regression","volume":"12","author":"Melly","year":"2005","journal-title":"Labour Econ"},{"key":"2023111615233105100_asad017-B33","first-page":"7322","article-title":"Counterfactual mean embeddings","volume":"22","author":"Muandet","year":"2021","journal-title":"The Journal of Machine Learning Research"},{"key":"2023111615233105100_asad017-B34","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1016\/j.jspi.2005.12.008","article-title":"Nonparametric causal effects based on marginal structural models","volume":"137","author":"Neugebauer","year":"2007","journal-title":"J. Statist. Plan. Infer"},{"key":"2023111615233105100_asad017-B35","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1307\/mmj\/1028999029","article-title":"Rational approximation to","volume":"11","author":"Newman","year":"1964","journal-title":"Michigan Math. J"},{"key":"2023111615233105100_asad017-B36","doi-asserted-by":"crossref","first-page":"399","DOI":"10.1016\/S0167-9473(97)00013-3","article-title":"Estimating the square root of a density via compactly supported wavelets","volume":"25","author":"Pinheiro","year":"1997","journal-title":"Comp. Statist. Data Anal"},{"key":"2023111615233105100_asad017-B37","volume-title":"R: A Language and Environment for Statistical Computing","author":"R Development Core Team","year":"2023"},{"key":"2023111615233105100_asad017-B38","doi-asserted-by":"crossref","first-page":"789","DOI":"10.3150\/14-BEJ679","article-title":"Empirical entropy, minimax regret and minimax risk","volume":"23","author":"Rakhlin","year":"2017","journal-title":"Bernoulli"},{"key":"2023111615233105100_asad017-B39","first-page":"547","volume-title":"Proc. 4th Berkeley Symp. Math. Statist. Prob","author":"R\u00e9nyi","year":"1961"},{"key":"2023111615233105100_asad017-B40","doi-asserted-by":"crossref","first-page":"260","DOI":"10.3103\/S1066530707030052","article-title":"Linear and convex aggregation of density estimators","volume":"16","author":"Rigollet","year":"2007","journal-title":"Math. Meth. Statist"},{"key":"2023111615233105100_asad017-B41","doi-asserted-by":"crossref","first-page":"2678","DOI":"10.1214\/10-AOS797","article-title":"Generalized density clustering","volume":"38","author":"Rinaldo","year":"2010","journal-title":"Ann. Statist."},{"key":"2023111615233105100_asad017-B42","doi-asserted-by":"crossref","first-page":"1951","DOI":"10.1214\/16-AOS1515","article-title":"Minimax estimation of a functional on a structured high dimensional model","volume":"45","author":"Robins","year":"2017","journal-title":"Ann. Statist"},{"key":"2023111615233105100_asad017-B43","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1214\/193940307000000527","volume-title":"Probability and Statistics: Essays in Honor of David A. Freedman","author":"Robins","year":"2008"},{"key":"2023111615233105100_asad017-B44","first-page":"920","article-title":"Comments on: Inference for semiparametric models: some questions and an answer","volume":"11","author":"Robins","year":"2001","journal-title":"Statist. Sinica"},{"key":"2023111615233105100_asad017-B45","doi-asserted-by":"crossref","first-page":"1305","DOI":"10.1214\/09-EJS479","article-title":"Semiparametric minimax rates","volume":"3","author":"Robins","year":"2009","journal-title":"Electron. J. Statist."},{"key":"2023111615233105100_asad017-B46","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1016\/j.jeconom.2009.09.001","article-title":"Nonparametric estimation of distributional policy effects","volume":"155","author":"Rothe","year":"2010","journal-title":"J. Economet"},{"key":"2023111615233105100_asad017-B47","doi-asserted-by":"crossref","first-page":"5973","DOI":"10.1109\/TIT.2016.2603151","article-title":"f-divergence inequalities","volume":"62","author":"Sason","year":"2016","journal-title":"IEEE Trans. Info. Theory"},{"key":"2023111615233105100_asad017-B48","doi-asserted-by":"crossref","first-page":"264","DOI":"10.1093\/ectj\/utaa027","article-title":"Debiased machine learning of conditional average treatment effects and other causal functions","volume":"24","author":"Semenova","year":"2021","journal-title":"Economet. J"},{"key":"2023111615233105100_asad017-B49","first-page":"303","volume-title":"Learning Theory and Kernel Machines, Lecture Notes Comp. Sci. 2777","author":"Tsybakov","year":"2003"},{"key":"2023111615233105100_asad017-B50","doi-asserted-by":"crossref","DOI":"10.1007\/b13794","volume-title":"Introduction to Nonparametric Estimation","author":"Tsybakov","year":"2009"},{"key":"2023111615233105100_asad017-B51","doi-asserted-by":"publisher","DOI":"10.2202\/1557-4679.1008","article-title":"Statistical inference for variable importance","volume":"2","author":"van der Laan","year":"2006","journal-title":"Int. J. Biostatistics"},{"key":"2023111615233105100_asad017-B52","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-21700-0","volume-title":"Unified Methods for Censored Longitudinal Data and Causality","author":"van der Laan","year":"2003"},{"key":"2023111615233105100_asad017-B53","volume-title":"Asymptotic Statistics","author":"van der Vaart","year":"2000"},{"key":"2023111615233105100_asad017-B54","first-page":"331","volume-title":"Lectures on Probability Theory and Statistics, Lecture Notes Math. 1781","author":"van der Vaart","year":"2002"},{"key":"2023111615233105100_asad017-B55","doi-asserted-by":"crossref","first-page":"1243","DOI":"10.1080\/01621459.2017.1330204","article-title":"Quantile-optimal treatment regimes","volume":"113","author":"Wang","year":"2018","journal-title":"J. Am. Statist. Assoc"},{"key":"2023111615233105100_asad017-B56","doi-asserted-by":"crossref","first-page":"2778","DOI":"10.1016\/j.jspi.2010.03.044","article-title":"Empirical likelihood confidence bands for distribution functions with missing responses","volume":"140","author":"Wang","year":"2010","journal-title":"J. Statist. Plan. Infer"},{"key":"2023111615233105100_asad017-B57","volume-title":"All of Nonparametric Statistics","author":"Wasserman","year":"2006"},{"key":"2023111615233105100_asad017-B58","doi-asserted-by":"crossref","first-page":"1001","DOI":"10.1214\/19-AOS1835","article-title":"A unified study of nonparametric inference for monotone functions","volume":"48","author":"Westling","year":"2020","journal-title":"Ann. Statist"},{"key":"2023111615233105100_asad017-B59","doi-asserted-by":"crossref","first-page":"149","DOI":"10.2307\/2526245","article-title":"Using least squares to approximate unknown regression functions","volume":"21","author":"White","year":"1980","journal-title":"Int. Econ. Rev"},{"key":"2023111615233105100_asad017-B60","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2307\/1912526","article-title":"Maximum likelihood estimation of misspecified models","volume":"50","author":"White","year":"1982","journal-title":"Econometrica"},{"key":"2023111615233105100_asad017-B61","volume-title":"Estimation, Inference and Specification Analysis","author":"White","year":"1996"},{"key":"2023111615233105100_asad017-B62","author":"Williamson","year":"2021"},{"key":"2023111615233105100_asad017-B63","author":"Wright","year":"2018"},{"key":"2023111615233105100_asad017-B64","doi-asserted-by":"crossref","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"}],"container-title":["Biometrika"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biomet\/advance-article-pdf\/doi\/10.1093\/biomet\/asad017\/50607309\/asad017.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/110\/4\/875\/53471755\/asad017.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/110\/4\/875\/53471755\/asad017.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,10,15]],"date-time":"2024-10-15T16:59:04Z","timestamp":1729011544000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biomet\/article\/110\/4\/875\/7068801"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,3]]},"references-count":64,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2023,3,3]]},"published-print":{"date-parts":[[2023,11,15]]}},"URL":"https:\/\/doi.org\/10.1093\/biomet\/asad017","relation":{},"ISSN":["0006-3444","1464-3510"],"issn-type":[{"value":"0006-3444","type":"print"},{"value":"1464-3510","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,12,1]]},"published":{"date-parts":[[2023,3,3]]}}}