An Introduction to the Augmented Inverse Propensity Weighted Estimator
The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
In this paper, we discuss an estimator for average treatment effects (ATEs) known as the augmented inverse propensity weighted (AIPW) estimator. This estimator has attractive theoretical properties and only requires practitioners to do two things they are already comfortable with: (1) specify a binary regression model for the propensity score, and (2) specify a regression model for the outcome variable. Perhaps the most interesting property of this estimator is its so-called “double robustness.” Put simply, the estimator remains consistent for the ATE if either the propensity score model or the outcome regression is misspecified but the other is properly specified. After explaining the AIPW estimator, we conduct a Monte Carlo experiment that compares the finite sample performance of the AIPW estimator to three common competitors: a regression estimator, an inverse propensity weighted (IPW) estimator, and a propensity score matching estimator. The Monte Carlo results show that the AIPW estimator has comparable or lower mean square error than the competing estimators when the propensity score and outcome models are both properly specified and, when one of the models is misspecified, the AIPW estimator is superior.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“These doubly robust approaches provide consistent estimates even if one of the models—the propensity score model or the outcome regression model—is misspecified.”
Citations
Cites
The central role of the propensity score in observational studies for causal effects
ROSENBAUM, P. R., & RUBIN, D. B. (1983) · Biometrika · doi:10.1093/biomet/70.1.41 · Cited by 31,241 · cited 2×
“Furthermore, we assume that potential outcomes are defined as in Rosenbaum and Rubin (1983) such that Yi(1) is the outcome that wewould observe if unit i had received treatment and Yi(0) is the outcome that we would observe if unit i had received treatment.”
BackgroundInfluentialCausality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653
CAUSALITY: MODELS, REASONING AND INFERENCE
Hitchcock, C. (2001) · The Philosophical Review · doi:10.1215/00318108-110-4-639 · Cited by 4,741
Identification of Causal Effects Using Instrumental Variables
Angrist, J. D., Imbens, G. W., & Rubin, D. B. (1996) · Journal of the American Statistical Association · doi:10.1080/01621459.1996.10476902 · Cited by 4,230
Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference
Ho, D. E., Imai, K., King, G., & Stuart, E. A. (2007) · Political Analysis · doi:10.1093/pan/mpl013 · Cited by 4,205 · cited 1×
“2 for a proof of this result and see Ho et al. (2007) for a related but distinct estimator that has this double robustness property.”
BackgroundHastie, T., & Tibshirani, R. (1986) · Statistical Science · doi:10.1214/ss/1177013604 · Cited by 3,409
Nonparametric Estimation of Average Treatment Effects Under Exogeneity: A Review
Imbens, G. W. (2004) · Review of Economics and Statistics · doi:10.1162/003465304323023651 · Cited by 2,994
Causality: models, reasoning, and inference
(2000) · Choice Reviews Online · doi:10.5860/choice.38-1494a · Cited by 2,873
Estimation of Regression Coefficients When Some Regressors are not Always Observed
Robins, J. M., Rotnitzky, A., & Zhao, L. P. (1994) · Journal of the American Statistical Association · doi:10.1080/01621459.1994.10476818 · Cited by 2,823 · cited 2×
“See Rubin (2006) for a book length treatment on matching, or Diamond and Sekhon (2005) and Ho et al.”
BackgroundRobins, J. (1986) · Mathematical Modelling · doi:10.1016/0270-0255(86)90088-6 · Cited by 2,451 · cited 2×
“Given the stated assumptions of this paper, it has been shown that such a model can be used to identify the ATE through a subclassification adjustment (Cochran 1968), the g-functional (Robins 1986), or the backdoor adjustment (Pearl 1995, 2000).”
BackgroundMethodInfluential
Research works printed but not linked: 4
- Busso, 2009a, Finite sample properties of semiparametric estimators of average treatment effects
- Sekhon, 2009, Multivariate and propensity score matching with balance optimization
- 2007, R: A language and environment for statistical computing
- Robins James M. 1999. Robust estimation in sequentially ignorable missing data and causal inference models. Proceedings of the American Statistical Association Section on Bayesian Statistical Science 6–10.
Cited by
Items 101–200 of 375
Suk, Y., & Kang, H. (2021) · Psychometrika · doi:10.1007/s11336-021-09805-x · Cited by 16
“If U ( 2 ) j were observed, a doubly robust estimator of the ATE (Bang & Robins, 2005; Glynn & Quinn, 2010) is: (4) In particular, the estimator ˆ τ is consistent for the ATE as long as the propensity score model or the outcome model is correctly specified, but not necessarily both.”
MethodCunningham, D. J., LaRose, M. A., Zhang, G. X., Au, S., MacAlpine, E. M., Paniagua, A. R., Klifto, C. S., & Gage, M. J. (2021) · Journal of Shoulder and Elbow Surgery · doi:10.1016/j.jse.2021.08.005 · Cited by 15
Implications of caesarean section for children's school achievement: A population‐based study
Smithers, L. G., Mol, B. W., Wilkinson, C., & Lynch, J. W. (2016) · Australian and New Zealand Journal of Obstetrics and Gynaecology · doi:10.1111/ajo.12475 · Cited by 15
Cesarean birth is not associated with early childhood body mass index
Smithers, L. G., Mol, B. W., Jamieson, L., & Lynch, J. (2016) · Pediatric Obesity · doi:10.1111/ijpo.12180 · Cited by 15
“Associations were analysed using augmented inverse probability of treatment weighting (aipw) (9,10).”
MethodCredit for households in Ghana: Has mobile money (momo) improved inclusive access?
Atta‐Ankomah, R. (2022) · Scientific African · doi:10.1016/j.sciaf.2022.e01230 · Cited by 14
Yang, M.-J., Chen, N., Ye, C.-Y., Li, Q., Luo, H., Wu, J., Liu, X., Guo, Q., Sessler, D. I., & Wang, E. (2024) · Journal of Clinical Anesthesia · doi:10.1016/j.jclinane.2024.111493 · Cited by 14
Farmland lease, high-rent threat and contract instability: evidence from China
Zhu, W., Paudel, K. P., Inoue, S., & Bi-liang, L. (2021) · China Agricultural Economic Review · doi:10.1108/caer-06-2020-0142 · Cited by 15
Unpacking The Nexus Between Broiler Contract Farming and Its Impact in Ghana
Bannor, R. K., & Gyekye, Y. (2022) · European Journal of Development Research · doi:10.1057/s41287-021-00489-8 · Cited by 14
“Further, the Rosenbaum Sensitivity analysis was deployed to analyse the robustness of the PSM estimates.”
BackgroundMethodInfluentialBettega, F., Leyrat, C., Tamisier, R., Mendelson, M., Grillet, Y., Sapène, M., Bonsignore, M. R., Pépin, J., Kattan, M. W., & Bailly, S. (2022) · Annals of the American Thoracic Society · doi:10.1513/annalsats.202109-1036oc · Cited by 17
Interrupting Inequality Through Community Land Trusts
Schneider, J. K., Lennon, M. C., & Saegert, S. (2022) · Housing Policy Debate · doi:10.1080/10511482.2022.2055614 · Cited by 19
“AIPW estimators combine models for the outcome and treatment status (Glynn & Quinn, 2010; Rubin & Laan, 2008; Tan, 2010; Wooldridge, 2007).”
MethodEffect of the Jamaica Early Childhood Stimulation Intervention on Labor Market Outcomes at Age 31
Heckman, J. J., Gertler, P., Pinto, R. R., Chang-Lopez, S. M., Grantham‐McGregor, S., Walker, S. P., Vermeersch, C. M. J., & Wright, A. S. (2021) · World Bank, Washington, DC eBooks · doi:10.1596/1813-9450-9787 · Cited by 15
The digital skill premium: Evidence from job vacancy data
García-Lazaro, A., Méndez-Astudillo, J., Lattanzio, S., Larkin, C. J., & Newnes, L. (2025) · Economics Letters · doi:10.1016/j.econlet.2025.112294 · Cited by 12
Continued Homelessness and Depressive Symptoms in Older Adults
Dobbins, S. K., Garcia, C. M., Evans, J. L., Valle, K. A., Guzman, D., & Kushel, M. (2024) · JAMA Network Open · doi:10.1001/jamanetworkopen.2024.27956 · Cited by 13
When does adjusting covariate under randomization help? A comparative study on current practices
Gao, Y., Liu, Y., & Matsouaka, R. (2024) · BMC Medical Research Methodology · doi:10.1186/s12874-024-02375-3 · Cited by 19
“…paper, we also consider the combination of the outcome regression and PS weighting methods to adjust for covariate imbalance— by using the augmented inverse probability weighting (AIPW) and the augmented overlap weighting (AOW) estimators [26–28]—as a natural extension of these separate methods.”
MethodHorsager, C., Færk, E., Gearhardt, A. N., Østergaard, S. D., & Lauritsen, M. B. (2021) · Eating and Weight Disorders - Studies on Anorexia Bulimia and Obesity · doi:10.1007/s40519-021-01113-8 · Cited by 14
Invariant Representation Learning for Treatment Effect EstimationPreprint
Shi, C., Veitch, V., & Blei, D. M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.12379 · Cited by 18
“There is extensive literature on different statistical estimators (Austin, 2011; Glynn and Quinn, 2010; VanderWeele and Shpitser, 2011; Funk et al., 2011) and machine learning methods adapted for causal inference (Hill, 2011; Athey and Imbens, 2016; Beck et al.”
BackgroundCunningham, D. J., Paniagua, A. R., LaRose, M. A., DeLaura, I. F., Blatter, M., & Gage, M. J. (2021) · Archives of Orthopaedic and Trauma Surgery · doi:10.1007/s00402-021-03892-2 · Cited by 11
Does noncompliance with COVID-19 regulations impact the depressive symptoms of others?
Oyenubi, A., & Kollamparambil, U. (2023) · Economic Modelling · doi:10.1016/j.econmod.2023.106191 · Cited by 13
Microfinance and poverty reduction: Evidence from Djibouti
Ali, M. A., Mughal, M., & Chhorn, D. (2022) · Review of Development Economics · doi:10.1111/rode.12856 · Cited by 16
Rodriguez, G. C., Bodnar, L. M., Brooks, M. M., Wahed, A. S., Kennedy, E. H., Schisterman, E. F., & Naimi, A. I. (2021) · American Journal of Epidemiology · doi:10.1093/aje/kwab220 · Cited by 16
Chlond, B., Gavard, C., & Jeuck, L. (2023) · Environmental and Resource Economics · doi:10.1007/s10640-022-00754-2 · Cited by 18
“Second, the adjustment term stabilizes the estimator when estimated propensity scores are close to 1 or 0, curing one undesirable property of the simple IPW estimator (see Glynn and Quinn 2010, for formal proof).”
BackgroundMethodInfluentialthe Austrian MS Treatment Registry (AMSTR), Guger, M., Enzinger, C., Leutmezer, F., Kraus, J., Kalcher, S., Kvas, E., & Berger, T. (2020) · Journal of Neurology · doi:10.1007/s00415-020-09811-6 · Cited by 11
Baker, S. E., Monlezun, D. J., Ambroze, W. L., & Margolin, D. A. (2020) · The American Surgeon · doi:10.1177/0003134820973720 · Cited by 13
Bannor, R. K., Amfo, B., Sarquah, K., Oppong-Kyeremeh, H., & Kyire, S. K. C. (2022) · International Journal of Energy Sector Management · doi:10.1108/ijesm-01-2022-0004 · Cited by 11
Horsager, C., LaFata, E. M., Færk, E., Lauritsen, M. B., Østergaard, S. D., & Gearhardt, A. N. (2023) · European Eating Disorders Review · doi:10.1002/erv.2974 · Cited by 10
Pontines, V. C. (2018) · Journal of Macroeconomics · doi:10.1016/j.jmacro.2018.06.007 · Cited by 11
Greenfield, E. A., Akıncıgil, A., & Moorman, S. M. (2019) · The Journals of Gerontology Series B · doi:10.1093/geronb/gbz132 · Cited by 12
Bannor, R. K., Oppong-Kyeremeh, H., Amfo, B., Hope, L., & Kyire, S. K. C. (2022) · SAGE Open · doi:10.1177/21582440221108170 · Cited by 10
“Regardless of these advantages over most impact assessment models, AIPW was less used among social scientists (Glynn & Quinn, 2010) and in particular, FBS impact studies.”
Heterogeneity within the Oregon Health Insurance Experiment: An application of causal forests
Hattab, Z., Doherty, E., Ryan, A. M., & O’Neill, S. (2024) · PLoS ONE · doi:10.1371/journal.pone.0297205 · Cited by 9
Guger, M., Enzinger, C., Leutmezer, F., Pauli, F. D., Kraus, J., Kalcher, S., Kvas, E., Berger, T., & Austrian MS Treatment Registry (AMSTR) (2024) · Journal of Neurology · doi:10.1007/s00415-024-12256-w · Cited by 14
“Thus, the misspecification of only one of the two models would not cause any harm to the ATE estimator [12].”
BackgroundErichsen, K., Šaras, E. D., & Perez‐Felkner, L. (2024) · Frontiers in Education · doi:10.3389/feduc.2024.1328574 · Cited by 13
Migrant remittances and the adoption of information and communication technology
Ali, U., Mughal, M., Ayaz, M., & Ahmed, J. (2024) · Information Economics and Policy · doi:10.1016/j.infoecopol.2024.101101 · Cited by 11
Spot check on corporate disclosure quality: Insights from policy learning
Zhang, J., Peng, Z., Zeng, Y., & Yang, H. (2024) · Pacific-Basin Finance Journal · doi:10.1016/j.pacfin.2024.102459 · Cited by 9
Ojo, T. O., Adesiyan, O. F., Ige, A. O., Emenike, C. U., Kassem, H. S., Elhindi, K. M., Kitole, F. A., & Sesabo, J. K. (2024) · Frontiers in Sustainable Food Systems · doi:10.3389/fsufs.2024.1414243 · Cited by 18
Gidelew, G. E., Alemu, B. A., & Kassie, K. E. (2024) · Journal of Agriculture and Food Research · doi:10.1016/j.jafr.2024.101620 · Cited by 15
Chow, V., & Vivalt, E. (2021) · Journal of African Economies · doi:10.1093/jae/ejab010 · Cited by 14
Cunningham, D. J., LaRose, M. A., Zhang, G. X., Paniagua, A. R., Klifto, C. S., & Gage, M. J. (2021) · Shoulder & Elbow · doi:10.1177/17585732211048117 · Cited by 9
“Generalized linear modeling was then carried out incorporating the propensity score weights, allowing for a “doubly robust” analysis.(15,16) Incident rate ratios from adjusted modeling were displayed, but treatment effect was then simulated for each patient within the dataset using model parameters to provide a median…”
MethodLoh, W. W., & Ren, D. (2023) · Psychological Methods · doi:10.1037/met0000564 · Cited by 9
Lesko, C. R., Falade-Nwulia, O. O., Pytell, J. D., Hutton, H. E., Fojo, A. T., Keruly, J. C., Moore, R. D., & Chander, G. (2023) · Addiction · doi:10.1111/add.16301 · Cited by 12
Naimi, A. I., Yu, Y., & Bodnar, L. M. (2024) · Epidemiology · doi:10.1097/ede.0000000000001785 · Cited by 16
“However, in our simulations, there were seed values that resulted in risk differences beyond this range (a known limitation of the augmented IPW estimator 24 ).”
Comparison of Survival After Aortic Valve Replacement With Mitroflow or Perimount Prostheses
Aasbjerg, K., Mortensen, P. E., Nørgaard, M. A., Rytgaard, H. C. W., Gerds, T. A., Søgaard, P., Torp-Pedersen, C. T., Mortensen, R. N., Bagge, B. J., Køber, L., et al. (2018) · Seminars in Thoracic and Cardiovascular Surgery · doi:10.1053/j.semtcvs.2018.11.016 · Cited by 9
Too young for Cannabis? Choice of minimum legal age for legalized non-medical Cannabis in Canada
Nguyen, H. V., Bornstein, S., Gamble, J., Mathews, M., Bishop, L. D., & Mital, S. (2020) · BMC Public Health · doi:10.1186/s12889-020-08639-z · Cited by 10
“It has been shown to outperform three popular estimators: a regression estimator, an inverse propensity weighted (IPW) estimator, and a propensity score matching estimator [30].”
BackgroundEvaluating multiple surrogate markers with censored data
Parast, L. M., Cai, T., & Tian, L. (2020) · Biometrics · doi:10.1111/biom.13370 · Cited by 11
“To derive a more robust estimator, we borrow ideas from the extensive causal inference literature focusing on double robust estimators (Lunceford and Davidian, 2004; Bang and Robins, 2005; Glynn and Quinn, 2010; Lee et al., 2010; Rotnitzky et al., 2012).”
MethodEmpowering women through microcredit in Djibouti
Ali, M. A., Mughal, M., & Chhorn, D. (2024) · Review of Development Economics · doi:10.1111/rode.13091 · Cited by 8
Ellul, S., Vansteelandt, S., Carlin, J. B., & Moreno‐Betancur, M. (2025) · Statistics in Medicine · doi:10.1002/sim.70272 · Cited by 19
“As promising alternatives to SR methods, Augmented Inverse Probability Weighting (AIPW) [15–17] and Targeted Maximum Likelihood Estimation (TMLE) [9, 18, 19] involve the fitting of both an outcome and an exposure model.”
Gender and racial price disparities in the NFT marketplace
Yuan, Y., Liu, X., Zhang, S., & Srinivasan, K. (2024) · International Journal of Research in Marketing · doi:10.1016/j.ijresmar.2024.08.008 · Cited by 8
Associations between Apgar scores and children’s educational outcomes at eight years of age
Yisma, E., Mol, B. W., Lynch, J. W., Mittinty, M., & Smithers, L. G. (2020) · Australian and New Zealand Journal of Obstetrics and Gynaecology · doi:10.1111/ajo.13220 · Cited by 12
Yisma, E., Mol, B. W., Lynch, J. W., Mittinty, M., & Smithers, L. G. (2020) · Ultrasound in Obstetrics and Gynecology · doi:10.1002/uog.23141 · Cited by 10
Regional anesthesia does not decrease opioid demand in pelvis and acetabulum fracture surgery
Cunningham, D. J., Robinette, J. P., Paniagua, A. R., LaRose, M. A., Blatter, M., & Gage, M. J. (2021) · European Journal of Orthopaedic Surgery & Traumatology · doi:10.1007/s00590-021-03114-w · Cited by 8
“Generalized linear modeling including propensity score weighting and the factors listed above was then carried out in a “doubly robust” fashion [43, 44].”
MethodShort supply chain choice and impact amongst rice processors in rural Ghana
Bannor, R. K., Oppong-Kyeremeh, H., Boateng, A. O., Bold, E., & Gruzah, B. (2023) · Journal of Agribusiness in Developing and Emerging Economies · doi:10.1108/jadee-03-2023-0049 · Cited by 8
Statistical Methods in Epidemiology
Ng, H. K. T., & Lovrić, M. (2025) · Springer eBooks · doi:10.1007/978-3-662-69359-9_616 · Cited by 10
The Relationship Between Work During College and Post College Earnings
Douglas, D., & Attewell, P. A. (2019) · Frontiers in Sociology · doi:10.3389/fsoc.2019.00078 · Cited by 20
“One important advantage of the AIPW technique is that if either the treatment model or the outcome model is incorrectly specified that the method nevertheless yields unbiased estimates of treatment, what statisticians call a “doubly robust” measure (Lunceford and Davidian, 2004; Funk et al., 2010; Glynn and Quinn, 201…”
MethodWang, B. (2025) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-04490-y · Cited by 8
“This robustness and minimisation of variability make AIPW a superior alternative to the augmented inverse propensity weighted (Glynn and Quinn 2010).”
Implementing TMLE in the presence of a continuous outcome
Frank, H. A., & Karim, M. E. (2023) · Research Methods in Medicine & Health Sciences · doi:10.1177/26320843231176662 · Cited by 7
“38 Despite this, TMLE has been shown to provide less biased estimates than AIPW since TMLE is more robust to near positivity violations and TMLE is a plug-in estimator restricting probabilities to be between 0 and 1.”
MethodKim, S., Bowman, N. A., Ingleby, L. D., Ford, D. C., & Sibaouih, C. (2018) · Educational Policy · doi:10.1177/0895904818813301 · Cited by 11
“…for this issue to be problematic, we used doubly robust estimation of causal treatment effects, which obtains unbiased results even when the analysis predicting either the treatment or the outcome—but not both—is misspecified (see Funk et al., 2011; Glynn & Quinn, 2010; Lunceford & Davidian, 2004).”
MethodPrague, M., Wang, R., & Victor, D. G. (2017) · The R Journal · doi:10.32614/rj-2017-041 · Cited by 7
“These methods have been extended to deal with missing data with IPW approaches CausalGAM (Glynn and Quinn 2010), but these packages are intended for analysis of observational studies, not CRTs.”
Quantifying the bias due to observed individual confounders in causal treatment effect estimates
Parast, L. M., & Griffin, B. A. (2020) · Statistics in Medicine · doi:10.1002/sim.8549 · Cited by 7
Opening Up: Capital Flows and Financial Sector Dynamics in Low-Income Developing Countries
Horn, S., & Narita, F. (2021) · IMF Working Paper · doi:10.5089/9781513597720.001 · Cited by 10
Landsittel, D. P., Chang, C. H., & Morton, S. C. (2019) · doi:10.25302/11.2019.me.130603827 · Cited by 7
Using Individualized Treatment Effects to Assess Treatment Effect Heterogeneity
Sechidis, K., Zhang, C., Sun, S. B., Chen, Y., Spector, A., & Bornkamp, B. (2025) · Statistics in Medicine · doi:10.1002/sim.70324 · Cited by 12
“This literature offers principled frameworks for estimating the ATE, with three of the most commonly used being G-computation, inverse probability weighting (IPW), and augmented IPW (AIPW) [18, 19]: where ̂𝜇 1 ( x 𝑖 ) is the estimated conditional expectation of the outcome given x 𝑖 within the…”
Effect of Internet usage on nonfarm employment participation and quality: Evidence from China
Liu, W., & Hu, Z. (2025) · Journal of Rural Studies · doi:10.1016/j.jrurstud.2025.103882 · Cited by 10
Kodalci, L., & Thas, O. (2023) · PLoS ONE · doi:10.1371/journal.pone.0292055 · Cited by 6
Pourat, N., Chen, X., Lu, C. M., Zhou, W., Yu-Lefler, H. F., Benjamin, T., Hoang, H., & Sripipatana, A. (2023) · Medical Care · doi:10.1097/mlr.0000000000001947 · Cited by 6
Improving Inverse Probability Weighting by Post-calibrating Its Propensity Scores
Gutman, R., Karavani, E., & Shimoni, Y. (2024) · Epidemiology · doi:10.1097/ede.0000000000001733 · Cited by 19
“It is very commonly used in methods for estimating causal effects from observational data, such as inverse probability weighting [1], propensity matching [2, 3], propensity stratification [4], as well as many doubly-robust methods [5, 6, 7, 8] Rosenbaum and Rubin [2] set up theoretical guaranties ensuring that adjusti…”
MethodA Primer on Deep Learning for Causal Inference
Koch, B., Sainburg, T., Geraldo, P., Jiang, S., Sun, Y., & Foster, J. G. (2024) · Sociological Methods & Research · doi:10.1177/00491241241234866 · Cited by 26
Cunningham, D. J., Paniagua, A. R., DeLaura, I. F., Zhang, G. X., Kim, B. I., Kim, J., Lee, T., LaRose, M. A., Adams, S. B., & Gage, M. J. (2022) · Foot & Ankle Specialist · doi:10.1177/19386400221088453 · Cited by 6
Papini, S., Chi, F. W., Schuler, A., Satre, D. D., Liu, V. X., & Sterling, S. (2022) · Drug and Alcohol Dependence · doi:10.1016/j.drugalcdep.2022.109607 · Cited by 7
Evaluating the impact of repeated leadership targeting on militant group durability
Tominaga, Y. (2019) · International Interactions · doi:10.1080/03050629.2019.1647836 · Cited by 6
Effect of conditional release on violent and general recidivism: A causal inference study
Karimi-Haghighi, M., Castillo, C., Tolan, S., & Lum, K. (2023) · Journal of Experimental Criminology · doi:10.1007/s11292-023-09596-4 · Cited by 12
“When the propensity score and outcome models are both properly specified and, when one of the models is misspecified, the AIPW estimator is superior (Glynn & Quinn, 2010).”
InfluentialEyasu, A. M., Zewotir, T. T., & Dessie, Z. G. (2024) · Scientific African · doi:10.1016/j.sciaf.2024.e02422 · Cited by 5
Butzin-Dozier, Z., Ji, Y., Coyle, J., Malenica, I., McQuade, E. T. R., Grembi, J. A., Platts-Mills, J. A., Houpt, E. R., Graham, J. P., Ali, S., et al. (2025) · PLoS neglected tropical diseases · doi:10.1371/journal.pntd.0012881 · Cited by 9
Using machine learning for efficient flexible regression adjustment in economic experiments
List, J. A., Muir, I., & Sun, G. (2024) · Econometric Reviews · doi:10.1080/07474938.2024.2373446 · Cited by 14
“Second, our approach generalizes a number of alternative practical implementations in the literature (Glynn and Quinn, 2010; Guo et al., 2021; Jin and Ba, 2023; Opper, 2021; Pitkin et al., 2013; Poyarkov et al., 2016; Rosenblum and Van Der Laan, 2010; Rothe, 2020; Spiess, 2018; Wager et al., 2016;…”
Ress, V., & Wild, E. (2024) · Health Economics · doi:10.1002/hec.4891 · Cited by 7
A Review of Causal Methods for High-Dimensional Data
Berkessa, Z. A., Läärä, E., & Waldmann, P. (2024) · IEEE Access · doi:10.1109/access.2024.3524261 · Cited by 9
Liao, L. D., Højbjerre‐Frandsen, E., Hubbard, A., & Schuler, A. (2025) · The International Journal of Biostatistics · doi:10.1515/ijb-2024-0018 · Cited by 24
“…machine learning internally to estimate the treatment or the outcome model, or both; for example, the augmented inverse probability weighting estimator (AIPW) and the targeted maximum likelihood estimator (TMLE) are commonly used to evaluate the average treatment effect [15, 16, 17, 18, 19].”
Guger, M., Enzinger, C., Leutmezer, F., Pauli, F. D., Kraus, J., Kalcher, S., Kvas, E., Berger, T., & the Austrian MS Treatment Registry (AMSTR) (2023) · Journal of Neurology · doi:10.1007/s00415-023-11644-y · Cited by 5
Atsebi, J.-M., Ligonnière, S., & Mathonnat, C. (2024) · Journal of International Money and Finance · doi:10.1016/j.jimonfin.2024.103220 · Cited by 6
Health shocks and health behavior: a long-term perspective
Bünnings, C., Simankova, I., & Tauchmann, H. (2025) · The European Journal of Health Economics · doi:10.1007/s10198-024-01747-2 · Cited by 5
“17 More specifically, we combine doubly robust treatment effect estimation, namely, augmented inverse-probability weighting [20, 27, 37], with machine learning techniques, double/debiased machine learning (DDML) [10, 11] in particular.”
InfluentialThe multi-store model for economic voting: Rome wasn't built in a day
Taniguchi, M. (2016) · Electoral Studies · doi:10.1016/j.electstud.2016.01.001 · Cited by 6
Introduction to the Virtual Issue: Past and Future Research Agenda on Causal Inference
Imai, K. (2011) · Political Analysis · doi:10.1017/s104719870001425x · Cited by 8
Evaluating sensitivity to classification uncertainty in latent subgroup effect analyses
Loh, W. W., & Kim, J. (2022) · BMC Medical Research Methodology · doi:10.1186/s12874-022-01720-8 · Cited by 11
“We will demonstrate the procedure using a doubly-robust augmented inverse propensity weighted (DR-AIPW) estimator [50, 51, 52] that is endowed with attractive statistical properties and combines (parametric regression) models for the outcome given treatment and covariates and for the treatment given covariates.”
MethodCentralized Deployment and Teacher Incentives: Evidence from Reforms in Rural China
Han, L., Liu, M., & An, X. (2016) · Economic Development and Cultural Change · doi:10.1086/689448 · Cited by 4
The Generalized Oaxaca-Blinder EstimatorPreprint
Guo, K., & Basse, G. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.11615 · Cited by 5
“e generalized Oaxaca-Blinder estimator is algebraically equivalent to the the augmented inverse-propensity weighted (AIPW) treatment eect estimator with known (and constant) treatment propensity (seeGlynn and Quinn (2010) for an overview). In the survey sampling literature, that estimator is known as the generalized d…”
MethodSallin, A. (2021) · RePEc: Research Papers in Economics · doi:10.13140/rg.2.2.20925.59365 · Cited by 12
Regression in tensor product spaces by the method of sieves
Zhang, T., & Simon, N. (2023) · Electronic Journal of Statistics · doi:10.1214/23-ejs2188 · Cited by 4
Methods in dynamic treatment regimens using observational healthcare data: A systematic review
Liang, D., Paul, A. K., Weir, D. L., Deneer, V. H. M., Greiner, R., Siebes, A. P. J. M., & Gardarsdóttir, H. (2025) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2025.108658 · Cited by 9
Does It Matter Who You Ask For Time Use Data?
Sharma, D., Swaminathan, H., & Lahoti, R. (2025) · The World Bank Economic Review · doi:10.1093/wber/lhaf004 · Cited by 10
Rate doubly robust estimation for weighted average treatment effects
Wang, Y., Liu, Y., & Yang, S. (2025) · Journal of Causal Inference · doi:10.1515/jci-2024-0073 · Cited by 13
“The augmented inverse probability weighted (AIPW) estimator for the ATE [12, 13], also known as the doubly robust (DR) estimator [11, 14, 15], ensures consistency if either the propensity score or outcome model is correctly speci fi ed.”
Robust location estimators in regression models with covariates and responses missing at random
Bianco, A. M., Boente, G., González–Manteiga, W., & Pérez-González, A. (2020) · Journal of nonparametric statistics · doi:10.1080/10485252.2020.1834108 · Cited by 4
“The proposed estimator is based in the augmented inverse probability weighting (aipw) method that was introduced, in the framework of casual inference, by Robins et al. (1994), Robbins (1999) and Scharfstein et al. (1999), see also Glynn and Quinn (2010).”
BackgroundMethodDo non-farmers pay more for land than farmers?
Schmidt, L., Odening, M., & Ritter, M. (2024) · European Review of Agricultural Economics · doi:10.1093/erae/jbae023 · Cited by 5
Special Economic Zones and Household Welfare: New Evidence from Ghana
Ackah, C., Osei, R., Owusu, N. Y. A., & Acheampong, V. (2025) · International Journal of the Economics of Business · doi:10.1080/13571516.2025.2456251 · Cited by 7
“…be consistent for the ATE whenever the propensity score model is correctly specified or the outcome regression is correctly specified, that is, it allows the estimator model for both the outcome and the treatment probability (Cattaneo, Drukker, and Holland 2013; Glynn and Quinn 2010; Kurz 2022).”
Liao, J., & Rohde, C. A. (2021) · Biometrics · doi:10.1111/biom.13454 · Cited by 10
“Other references of IPW estimators include Hirano et al. (2000), Glynn and Quinn (2010), Williamson et al. (2014), Austin and Stuart (2015), Imbens and Rubin (2015), and Li et al. (2018).”
BackgroundDecreased Susceptibility of Marginal Odds Ratios to Finite-sample Bias
Ross, R. K., Cole, S. R., & Richardson, D. B. (2021) · Epidemiology · doi:10.1097/ede.0000000000001370 · Cited by 5
Austin, P. C., Rubin, D. B., & Thomas, N. (2021) · Statistics in Medicine · doi:10.1002/sim.9141 · Cited by 4
Conservation agriculture's effect on smallholder farmer wellbeing in Mozambique
Daihawe, D. L., Lambert, D. M., Mulungu, K., & Eash, N. S. (2021) · Renewable Agriculture and Food Systems · doi:10.1017/s1742170521000430 · Cited by 4
The impact of regional anesthesia on opioid demand in distal radius fracture surgery
Cunningham, D. J., LaRose, M. A., Kinamon, T., MacAlpine, E. M., Au, S., Paniagua, A. R., Klifto, C. S., & Gage, M. J. (2022) · Journal of Plastic Surgery and Hand Surgery · doi:10.1080/2000656x.2022.2070178 · Cited by 4
“In this way, analyses were performed in a ‘doubly robust’ fashion [36,37].”
MethodStults, C. D., Yan, X. S., Deng, S., Dillon, E. C., Liang, S., Jones, J., Bhanushali, M. J., & Frosch, D. L. (2022) · Journal of Patient Experience · doi:10.1177/23743735221113160 · Cited by 5
Transfering Targeted Maximum Likelihood Estimation for Causal Inference into Sports Science
Dijkhuis, T. B., & Blaauw, F. J. (2022) · Entropy · doi:10.3390/e24081060 · Cited by 4
“) or the outcome model is misspecified [6].”
BackgroundRudolph, J. E., Schisterman, E. F., & Naimi, A. I. (2022) · American Journal of Epidemiology · doi:10.1093/aje/kwac162 · Cited by 8
Property management technology adoption in the short‐term housing rental market
Göppinger, S., Luque, J. P., & Marcato, G. (2024) · Real Estate Economics · doi:10.1111/1540-6229.12504 · Cited by 11
Metadata
- Authors
- Glynn, A. N., & Quinn, K. M.
- Affiliations
- Harvard University (US)
- University of California, Berkeley (US)
- Corresponding author
- Glynn, Adam N.
- Venue
- Political Analysis
- Publisher
- Cambridge University Press (CUP)
- Published
- 2010
- Volume
- 18
- Issue
- 1
- Pages
- 36-56
- ISSN
- 1047-1987 (ISSN-L)
- 1476-4989
- OpenAlex
- W2066443416
- Cited by
- 374
- References
- 33
- Citations by year
- 2026: 29 · 2025: 57 · 2024: 58 · 2023: 44 · 2022: 36 · 2021: 45 · 2020: 25 · 2019: 18 · 2018: 12 · 2017: 9 · 2016: 15 · 2015: 6 · 2014: 6 · 2013: 9 · 2012: 2
- FWCI
- 1.11
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Political Analysis (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Keywords
- average treatment effect, double robustness, propensity score, inverse propensity weighting, propensity score matching, model misspecification, finite sample properties, causal inference
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.