{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T22:32:43Z","timestamp":1793831563987,"version":"4.6.1"},"reference-count":46,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2022,10,29]],"date-time":"2022-10-29T00:00:00Z","timestamp":1667001600000},"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":[[2023,8,14]]},"abstract":"<jats:title>Summary<\/jats:title>\n               <jats:p>A basic principle in the design of observational studies is to approximate the randomized experiment that would have been conducted under ideal circumstances. At present, linear regression models are commonly used to analyse observational data and estimate causal effects. How do linear regression adjustments in observational studies emulate key features of randomized experiments, such as covariate balance, self-weighted sampling and study representativeness? In this paper, we provide answers to this and related questions by analysing the implied individual-level data weights of various linear regression methods. We derive new closed-form expressions of these implied weights, and examine their properties in both finite and large samples. Among others, in finite samples we characterize the implied target population of linear regression, and in large samples demonstrate the multiply robust properties of regression estimators from the perspective of their implied weights. We show that the implied weights of general regression methods can be equivalently obtained by solving a convex optimization problem. This equivalence allows us to bridge ideas from the regression modelling and causal inference literatures. As a result, we propose novel regression diagnostics for causal inference that are part of the design stage of an observational study. We implement the weights and diagnostics in the new lmw package for R.<\/jats:p>","DOI":"10.1093\/biomet\/asac058","type":"journal-article","created":{"date-parts":[[2022,10,29]],"date-time":"2022-10-29T13:38:40Z","timestamp":1667050720000},"page":"615-629","source":"Crossref","is-referenced-by-count":29,"title":["On the implied weights of linear regression for causal inference"],"prefix":"10.1093","volume":"110","author":[{"given":"Ambarish","family":"Chattopadhyay","sequence":"first","affiliation":[{"name":"Stanford Data Science, Stanford University , 450 Jane Stanford Way, Wallenberg Hall, Stanford, California 94305, U.S.A"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jos\u00e9 R","family":"Zubizarreta","sequence":"additional","affiliation":[{"name":"Harvard University Departments of Health Care Policy, Biostatistics, and Statistics, , 180 A Longwood Avenue, Boston, Massachusetts 02115, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,10,29]]},"reference":[{"key":"2023081420144664400_B1","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1016\/S0304-4076(02)00201-4","article-title":"Semiparametric instrumental variable estimation of treatment response models","volume":"113","author":"Abadie,","year":"2003","journal-title":"J. Economet."},{"key":"2023081420144664400_B2","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1111\/ajps.12116","article-title":"Comparative politics and the synthetic control method","volume":"59","author":"Abadie,","year":"2015","journal-title":"Am. J. Polit. Sci."},{"key":"2023081420144664400_B3","doi-asserted-by":"crossref","first-page":"249","DOI":"10.2307\/2998558","article-title":"Estimating the labor market impact of voluntary military service using social security data on military applicants","volume":"66","author":"Angrist,","year":"1998","journal-title":"Econometrica"},{"key":"2023081420144664400_B4","first-page":"1277","article-title":"Empirical strategies in labor economics","volume-title":"Handbook of Labor Economics","author":"Angrist,","year":"1999"},{"key":"2023081420144664400_B5","doi-asserted-by":"crossref","DOI":"10.2307\/j.ctvcm4j72","volume-title":"Mostly Harmless Econometrics: An Empiricist\u2019s Companion","author":"Angrist,","year":"2008"},{"key":"2023081420144664400_B6","doi-asserted-by":"crossref","first-page":"250","DOI":"10.1111\/ajps.12185","article-title":"Does regression produce representative estimates of causal effects?","volume":"60","author":"Aronow,","year":"2016","journal-title":"Am. J. Polit. Sci."},{"key":"2023081420144664400_B7","doi-asserted-by":"crossref","first-page":"1253","DOI":"10.1377\/hlthaff.2020.00598","article-title":"Disparities in outcomes among COVID-19 patients in a large health care system in California","volume":"39","author":"Azar,","year":"2020","journal-title":"Health Affairs"},{"key":"2023081420144664400_B8","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1080\/01621459.2021.1929245","article-title":"The augmented synthetic control method","volume":"116","author":"Ben-Michael,","year":"2021","journal-title":"J. Am. Statist. Assoc."},{"key":"2023081420144664400_B9","doi-asserted-by":"crossref","first-page":"3227","DOI":"10.1002\/sim.8659","article-title":"Balancing vs modeling approaches to weighting in practice","volume":"39","author":"Chattopadhyay,","year":"2020","journal-title":"Statist. Med."},{"key":"2023081420144664400_B10","first-page":"234","article-title":"The planning of observational studies of human populations","volume":"128","author":"Cochran,","year":"1965","journal-title":"J. R. Statist. Soc."},{"key":"2023081420144664400_B11","volume-title":"Residuals and Influence in Regression","author":"Cook,","year":"1982"},{"key":"2023081420144664400_B12","doi-asserted-by":"crossref","first-page":"2964","DOI":"10.1257\/aer.20181169","article-title":"Two-way fixed effects estimators with heterogeneous treatment effects","volume":"110","author":"De Chaisemartin,","year":"2020","journal-title":"Am. Econ. Rev."},{"key":"2023081420144664400_B13","doi-asserted-by":"crossref","first-page":"376","DOI":"10.1080\/01621459.1992.10475217","article-title":"Calibration estimators in survey sampling","volume":"87","author":"Deville,","year":"1992","journal-title":"J. Am. Statist. Assoc."},{"key":"2023081420144664400_B14","doi-asserted-by":"crossref","first-page":"108945","DOI":"10.1016\/j.spl.2020.108945","article-title":"The Frisch--Waugh--Lovell theorem for standard errors","volume":"168","author":"Ding,","year":"2021","journal-title":"Statist. Prob. Lett."},{"key":"2023081420144664400_B15","doi-asserted-by":"crossref","first-page":"605","DOI":"10.1017\/pan.2019.25","article-title":"A bracketing relationship between difference-in-differences and lagged-dependent-variable adjustment","volume":"27","author":"Ding,","year":"2019","journal-title":"Polit. Anal."},{"key":"2023081420144664400_B16","doi-asserted-by":"crossref","first-page":"677","DOI":"10.2105\/AJPH.43.6_Pt_1.677","article-title":"Philosophy of inferences from retrospective studies","volume":"43","author":"Dorn,","year":"1953","journal-title":"Am. J. Public Health Nations Health"},{"key":"2023081420144664400_B17","doi-asserted-by":"crossref","DOI":"10.1002\/9780470523551","volume-title":"Sampling Statistics","author":"Fuller,","year":"2009"},{"key":"2023081420144664400_B18","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1080\/07350015.2017.1366909","article-title":"Why high-order polynomials should not be used in regression discontinuity designs","volume":"37","author":"Gelman,","year":"2018","journal-title":"J. Bus. Econ. Statist."},{"key":"2023081420144664400_B19","article-title":"Contamination bias in linear regressions","author":"Goldsmith-Pinkham,","year":"2022","journal-title":"arXiv: 2106.05024v3"},{"key":"2023081420144664400_B20","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1093\/pan\/mpr025","article-title":"Entropy balancing for causal effects: a multivariate reweighting method to produce balanced samples in observational studies","volume":"20","author":"Hainmueller,","year":"2012","journal-title":"Polit. Anal."},{"key":"2023081420144664400_B21","volume-title":"Causal Inference: What If","author":"Hern\u00e1n,","year":"2020"},{"key":"2023081420144664400_B22","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1093\/pan\/mpl013","article-title":"Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference","volume":"15","author":"Ho,","year":"2007","journal-title":"Polit. Anal."},{"key":"2023081420144664400_B23","doi-asserted-by":"crossref","first-page":"467","DOI":"10.1111\/ajps.12417","article-title":"When should we use unit fixed effects regression models for causal inference with longitudinal data?","volume":"63","author":"Imai,","year":"2019","journal-title":"Am. J. Polit. Sci."},{"key":"2023081420144664400_B24","doi-asserted-by":"crossref","first-page":"373","DOI":"10.3368\/jhr.50.2.373","article-title":"Matching methods in practice: three examples","volume":"50","author":"Imbens,","year":"2015","journal-title":"J. Hum. Resour."},{"key":"2023081420144664400_B25","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9781139025751","volume-title":"Causal Inference in Statistics, Social, and Biomedical Sciences","author":"Imbens,","year":"2015"},{"key":"2023081420144664400_B26","first-page":"523","article-title":"Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data (with discussion)","volume":"22","author":"Kang,","year":"2007","journal-title":"Statist. Sci."},{"key":"2023081420144664400_B27","doi-asserted-by":"crossref","DOI":"10.1201\/9780429321740","volume-title":"Statistical Methods for Handling Incomplete Data","author":"Kim,","year":"2021"},{"key":"2023081420144664400_B28","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1093\/pan\/mpj004","article-title":"The dangers of extreme counterfactuals","volume":"14","author":"King,","year":"2006","journal-title":"Polit. Anal."},{"key":"2023081420144664400_B29","volume-title":"Survey Sampling","author":"Kish,","year":"1965"},{"key":"2023081420144664400_B30","doi-asserted-by":"crossref","first-page":"532","DOI":"10.1257\/aer.101.3.532","article-title":"Oaxaca-Blinder as a reweighting estimator","volume":"101","author":"Kline,","year":"2011","journal-title":"Am. Econ. Rev."},{"key":"2023081420144664400_B31","first-page":"604","article-title":"Evaluating the econometric evaluations of training programs with experimental data","volume":"76","author":"LaLonde,","year":"1986","journal-title":"Am. Econ. Rev."},{"key":"2023081420144664400_B32","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1214\/12-AOAS583","article-title":"Agnostic notes on regression adjustments to experimental data: reexamining Freedman\u2019s critique","volume":"7","author":"Lin,","year":"2013","journal-title":"Ann. Appl. Statist."},{"key":"2023081420144664400_B33","first-page":"1","article-title":"On the application of probability theory to agricultural experiments. Essay on principles. Section 9","volume":"X","author":"Neyman,","year":"1923","journal-title":"Ann. Agric. Sci."},{"key":"2023081420144664400_B34","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1146\/annurev-statistics-022513-115617","article-title":"League tables for hospital comparisons","volume":"3","author":"Normand,","year":"2016","journal-title":"Ann. Rev. Statist. Appl."},{"key":"2023081420144664400_B35","volume-title":"R: A Language and Environment for Statistical Computing","year":"2023"},{"key":"2023081420144664400_B36","doi-asserted-by":"crossref","first-page":"544","DOI":"10.1214\/07-STS227D","article-title":"Comment: performance of double-robust estimators when \u2018inverse probability\u2019 weights are highly variable","volume":"22","author":"Robins,","year":"2007","journal-title":"Statist. Sci."},{"key":"2023081420144664400_B37","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1080\/01621459.1994.10476818","article-title":"Estimation of regression coefficients when some regressors are not always observed","volume":"89","author":"Robins,","year":"1994","journal-title":"J. Am. Statist. Assoc."},{"key":"2023081420144664400_B38","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4419-1213-8","volume-title":"Design of Observational Studies","author":"Rosenbaum,","year":"2010"},{"key":"2023081420144664400_B39","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1093\/biomet\/70.1.41","article-title":"The central role of the propensity score in observational studies for causal effects","volume":"70","author":"Rosenbaum,","year":"1983","journal-title":"Biometrika"},{"key":"2023081420144664400_B40","doi-asserted-by":"crossref","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":"J. Educ. Psychol."},{"key":"2023081420144664400_B41","doi-asserted-by":"crossref","first-page":"591","DOI":"10.2307\/2287653","article-title":"Randomization analysis of experimental data: the Fisher randomization test comment","volume":"75","author":"Rubin,","year":"1980","journal-title":"J. Am. Statist. Assoc."},{"key":"2023081420144664400_B42","doi-asserted-by":"crossref","first-page":"808","DOI":"10.1214\/08-AOAS187","article-title":"For objective causal inference, design trumps analysis","volume":"2","author":"Rubin,","year":"2008","journal-title":"Ann. Appl. Statist."},{"key":"2023081420144664400_B43","article-title":"A general weighted average representation of the ordinary and two-stage least squares estimands","author":"S\u0142oczy\u0144ski,","year":"2018","journal-title":"IZA Discussion Papers, No. 11866"},{"key":"2023081420144664400_B44","doi-asserted-by":"crossref","first-page":"501","DOI":"10.1162\/rest_a_00953","article-title":"Interpreting OLS estimands when treatment effects are heterogeneous: smaller groups get larger weights","volume":"104","author":"S\u0142oczy\u0144ski,","year":"2020","journal-title":"Rev. Econ. Statist."},{"key":"2023081420144664400_B45","doi-asserted-by":"crossref","first-page":"910","DOI":"10.1080\/01621459.2015.1023805","article-title":"Stable weights that balance covariates for estimation with incomplete outcome data","volume":"110","author":"Zubizarreta,","year":"2015","journal-title":"J. Am. Statist. Assoc."},{"key":"2023081420144664400_B46","doi-asserted-by":"crossref","first-page":"204","DOI":"10.1214\/13-AOAS713","article-title":"Matching for balance, pairing for heterogeneity in an observational study of the effectiveness of for-profit and not-for-profit high schools in Chile","volume":"8","author":"Zubizarreta,","year":"2014","journal-title":"Ann. Appl. Statist."}],"container-title":["Biometrika"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/biomet\/advance-article-pdf\/doi\/10.1093\/biomet\/asac058\/48826850\/asac058.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/110\/3\/615\/51111540\/asac058.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/110\/3\/615\/51111540\/asac058.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,14]],"date-time":"2023-08-14T20:36:24Z","timestamp":1692045384000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biomet\/article\/110\/3\/615\/6779968"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,29]]},"references-count":46,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2022,10,29]]},"published-print":{"date-parts":[[2023,8,14]]}},"URL":"https:\/\/doi.org\/10.1093\/biomet\/asac058","relation":{},"ISSN":["0006-3444","1464-3510"],"issn-type":[{"value":"0006-3444","type":"print"},{"value":"1464-3510","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,9,1]]},"published":{"date-parts":[[2022,10,29]]}}}