Hostname: page-component-7857688df4-qjfxt Total loading time: 0 Render date: 2025-11-15T01:36:08.467Z Has data issue: false hasContentIssue false
  • English
  • Français

ISOTONIC PROPENSITY SCORE MATCHING

Published online by Cambridge University Press:  14 November 2025

Mengshan Xu  and
Taisuke Otsu
Mengshan Xu
Affiliation:
University of Mannheim
Taisuke Otsu*
Affiliation:
London School of Economics
*
Address correspondence to Taisuke Otsu, Department of Economics, London School of Economics, London, U.K., e-mail: t.otsu@lse.ac.uk.
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

We propose a one-to-many matching estimator of the average treatment effect based on propensity scores estimated by isotonic regression. This approach is predicated on the assumption of monotonicity in the propensity score function, a condition that can be justified in many economic applications. We show that the nature of the isotonic estimator can help us to fix many problems of existing matching methods, including efficiency, choice of the number of matches, choice of tuning parameters, robustness to propensity score misspecification, and bootstrap validity. As a by-product, a uniformly consistent isotonic estimator is developed for our proposed matching method.

Information

Type
ARTICLES
Information
Econometric Theory , First View , pp. 1 - 51
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Footnotes

We are grateful to Markus Frölich, Daniel Gutknecht, Phillip Heiler, Lihua Lei, Yoshi Rai, Christoph Rothe, Carsten Trenkler, and participants at the econometrics seminar at Mannheim 2022, NASMES 2023, and IAAE 2023, for helpful comments and discussions. We also would like to thank a co-editor and anonymous referees for helpful comments to revise the paper.

References

REFERENCES

Abadie, A., & Imbens, G. W. (2006). Large sample properties of matching estimators for average treatment effects. Econometrica , 74, 235–267.10.1111/j.1468-0262.2006.00655.xCrossRefGoogle Scholar
Abadie, A., & Imbens, G. W. (2008). On the failure of the bootstrap for matching estimators. Econometrica , 76, 1537–1557.Google Scholar
Abadie, A., & Imbens, G. W. (2011). Bias-corrected matching estimators for average treatment effects. Journal of Business & Economic Statistics , 29, 1–11.10.1198/jbes.2009.07333CrossRefGoogle Scholar
Abadie, A., & Imbens, G. W. (2016). Matching on the estimated propensity score. Econometrica , 84, 781–807.10.3982/ECTA11293CrossRefGoogle Scholar
Adusumilli, K. (2020). Bootstrap inference for propensity score matching. Working Paper.Google Scholar
Ai, C., & Chen, X. (2003). Efficient estimation of models with conditional moment restrictions containing unknown functions. Econometrica , 71, 1795–1843.10.1111/1468-0262.00470CrossRefGoogle Scholar
Andrews, D. W. K. (1994). Asymptotics for semiparametric econometric models via stochastic equicontinuity. Econometrica , 62, 43–72.10.2307/2951475CrossRefGoogle Scholar
Armstrong, T. B., & Kolesár, M. (2021). Finite-sample optimal estimation and inference on average treatment effects under unconfoundedness. Econometrica , 89, 1141–1177.10.3982/ECTA16907CrossRefGoogle Scholar
Ayer, M., Brunk, H. D., Ewing, G. M., Reid, W. T., & Silverman, E. (1955). An empirical distribution function for sampling with incomplete information. Annals of Mathematical Statistics , 26, 641–647.10.1214/aoms/1177728423CrossRefGoogle Scholar
Babii, A., & Kumar, R. (2023). Isotonic regression discontinuity designs. Journal of Econometrics , 234(2), 371–393.10.1016/j.jeconom.2021.01.008CrossRefGoogle Scholar
Balabdaoui, F., Durot, C., & Jankowski, H. (2019). Least squares estimation in the monotone single index model. Bernoulli , 25, 3276–3310.10.3150/18-BEJ1090CrossRefGoogle Scholar
Balabdaoui, F., Groeneboom, P., & Hendrickx, K. (2019). Score estimation in the monotone single index model. Scandinavian Journal of Statistics , 46, 517–544.10.1111/sjos.12361CrossRefGoogle Scholar
Balabdaoui, F., & Groeneboom, P. (2021). Profile least squares estimators in the monotone single index model. In A. Daouia and A. Ruiz-Gazen (Eds.), Advances in contemporary statistics and econometrics (pp. 3–22). Springer.10.1007/978-3-030-73249-3_1CrossRefGoogle Scholar
Barlow, R. E., Bartholomew, D. J., Bremner, J. M., & Brunk, H. D. (1972). Statistical inference under order restrictions: The theory and application of isotonic regression . John Wiley & Sons.Google Scholar
Barlow, R., & Brunk, H. (1972). The isotonic regression problem and its dual. Journal of the American Statistical Association , 67, 140–147.10.1080/01621459.1972.10481216CrossRefGoogle Scholar
Bickel, P. J., & Ritov, Y. A. (2003). Nonparametric estimators which can be “plugged-in. Annals of Statistics , 31, 1033–1053.10.1214/aos/1059655904CrossRefGoogle Scholar
Bodory, H., Camponovo, L., Huber, M., & Lechner, M. (2016). A wild bootstrap algorithm for propensity score matching estimators. Working Paper.Google Scholar
Brunk, H. D. (1958). On the estimation of parameters restricted by inequalities. Annals of Mathematical Statistics , 29, 437–454.10.1214/aoms/1177706621CrossRefGoogle Scholar
Cattaneo, M. D., & Farrell, M. H. (2013). Optimal convergence rates, bahadur representation, and asymptotic normality of partitioning estimators. Journal of Econometrics , 174, 127–143.10.1016/j.jeconom.2013.02.002CrossRefGoogle Scholar
Chamberlain, G. (1987). Asymptotic efficiency in estimation with conditional moment restrictions. Journal of Econometrics , 34, 305–334.10.1016/0304-4076(87)90015-7CrossRefGoogle Scholar
Chen, X., Linton, O., & Van Keilegom, I. (2003). Estimation of semiparametric models when the criterion function is not smooth. Econometrica , 71, 1591–1608.10.1111/1468-0262.00461CrossRefGoogle Scholar
Chen, X., & Santos, A. (2018). Overidentification in regular models. Econometrica , 86, 1771–1817.10.3982/ECTA13559CrossRefGoogle Scholar
Cheng, G. (2009). Semiparametric additive isotonic regression. Journal of Statistical Planning and Inference , 139, 1980–1991.10.1016/j.jspi.2008.09.009CrossRefGoogle Scholar
Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., & Newey, W. (2017). Double/debiased/Neyman machine learning of treatment effects. American Economic Review , 107, 261–265.10.1257/aer.p20171038CrossRefGoogle Scholar
Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal , 21, C1–C68.10.1111/ectj.12097CrossRefGoogle Scholar
Chernozhukov, V., Escanciano, J. C., Ichimura, H., Newey, W. K., & Robins, J. M. (2022). Locally robust semiparametric estimation. Econometrica , 90, 1501–1535.10.3982/ECTA16294CrossRefGoogle Scholar
Chernozhukov, V., Newey, W. K., & Singh, R. (2022). Automatic debiased machine learning of causal and structural effects. Econometrica , 90, 967–1027.10.3982/ECTA18515CrossRefGoogle Scholar
Cosslett, S. R. (1983). Distribution-free maximum likelihood estimator of the binary choice model. Econometrica , 51, 765–782.10.2307/1912157CrossRefGoogle Scholar
Cosslett, S. R. (1987). Efficiency bounds for distribution-free estimators of the binary choice and the censored regression models. Econometrica , 55, 559–585.10.2307/1913600CrossRefGoogle Scholar
Cosslett, S. R. (2007). Efficient estimation of semiparametric models by smoothed maximum likelihood. International Economic Review , 48, 1245–1272.10.1111/j.1468-2354.2007.00461.xCrossRefGoogle Scholar
Dehejia, R. H., & Wahba, S. (1999). Causal effects in nonexperimental studies: Reevaluating the evaluation of training programs. Journal of the American Statistical Association , 94, 1053–1062.10.1080/01621459.1999.10473858CrossRefGoogle Scholar
Dehejia, R. H., & Wahba, S. (2002). Propensity score-matching methods for nonexperimental causal studies. Review of Economics and Statistics , 84, 151–161.10.1162/003465302317331982CrossRefGoogle Scholar
Durot, C., Kulikov, V. N., & Lopuhaä, H. P. (2012). The limit distribution of the ${L}_{\infty }$ -error of Grenander-type estimators. Annals of Statistics , 40, 1578–1608.10.1214/12-AOS1015CrossRefGoogle Scholar
Frölich, M. (2004). Finite-sample properties of propensity-score matching and weighting estimators. Review of Economics and Statistics , 86, 77–90.10.1162/003465304323023697CrossRefGoogle Scholar
Frölich, M., Huber, M., & Wiesenfarth, M. (2017). The finite sample performance of semi-and non-parametric estimators for treatment effects and policy evaluation. Computational Statistics & Data Analysis , 115, 91–102.10.1016/j.csda.2017.05.007CrossRefGoogle Scholar
Grenander, U. (1956). On the theory of mortality measurement, II. Skand Aktuarietidskr , 39, 125–153.Google Scholar
Groeneboom, P., & Hendrickx, K. (2017). The nonparametric bootstrap for the current status model. Electronic Journal of Statistics , 11, 3446–3484.10.1214/17-EJS1345CrossRefGoogle Scholar
Groeneboom, P., & Hendrickx, K. (2018). Current status linear regression. Annals of Statistics , 46, 1415–1444.10.1214/17-AOS1589CrossRefGoogle Scholar
Groeneboom, P., & Jongbloed, G. (2014). Nonparametric estimation under shape constraints . Cambridge University Press.10.1017/CBO9781139020893CrossRefGoogle Scholar
Györfi, L., Kohler, M., Krzyzak, A., & Walk, H. (2002). A distribution-free theory of nonparametric regression (Vol. 1). Springer Science & Business Media.10.1007/b97848CrossRefGoogle Scholar
Hahn, J. (1998). On the role of the propensity score in efficient semiparametric estimation of average treatment effects. Econometrica , 66, 315–331.10.2307/2998560CrossRefGoogle Scholar
Han, A. K. (1987). Non-parametric analysis of a generalized regression model. Journal of Econometrics , 35, 303–316.10.1016/0304-4076(87)90030-3CrossRefGoogle Scholar
Heckman, J. J., Ichimura, H., & Todd, P. E. (1997). Matching as an econometric evaluation estimator: Evidence from evaluating a job training programme. Review of Economic Studies , 64, 605–654.10.2307/2971733CrossRefGoogle Scholar
Heckman, J. J., Ichimura, H., & Todd, P. (1998). Matching as an econometric evaluation estimator. Review of Economic Studies , 65, 261–294.10.1111/1467-937X.00044CrossRefGoogle Scholar
Heckman, J., Ichimura, H., Smith, J., & Todd, P. (1998). Characterizing selection bias using experimental data. Econometrica , 66, 1017–1098.10.2307/2999630CrossRefGoogle Scholar
Hirano, K., Imbens, G. W., & Ridder, G. (2003). Efficient estimation of average treatment effects using the estimated propensity score. Econometrica , 71, 1161–1189.10.1111/1468-0262.00442CrossRefGoogle Scholar
Huang, J. (2002). A note on estimating a partly linear model under monotonicity constraints. Journal of Statistical Planning and Inference , 107, 343–351.10.1016/S0378-3758(02)00262-8CrossRefGoogle Scholar
Imbens, G. W. (2004). Nonparametric estimation of average treatment effects under exogeneity: A review. Review of Economics and Statistics , 86, 4–29.10.1162/003465304323023651CrossRefGoogle Scholar
Khan, S., & Tamer, E. (2010). Irregular identification, support conditions, and inverse weight estimation. Econometrica , 78, 2021–2042.Google Scholar
Klein, R. W., & Spady, R. H. (1993). An efficient semiparametric estimator for binary response models. Econometrica , 61, 387–421.10.2307/2951556CrossRefGoogle Scholar
Lin, Z., Ding, P., & Han, F. (2023). Estimation based on nearest neighbor matching: From density ratio to average treatment effect. Econometrica , 91, 2187–2217.10.3982/ECTA20598CrossRefGoogle Scholar
Liu, Y., & Qin, J. (2022). Tuning-parameter-free optimal propensity score matching approach for causal inference. Preprint, arXiv:2205.13200.Google Scholar
Liu, Y., & Qin, J. (2024). Tuning-parameter-free propensity score matching approach for causal inference under shape restriction. Journal of Econometrics , 244, 105829.10.1016/j.jeconom.2024.105829CrossRefGoogle Scholar
Matzkin, R. L. (1992). Nonparametric and distribution-free estimation of the binary threshold crossing and the binary choice models. Econometrica , 60, 239–270.10.2307/2951596CrossRefGoogle Scholar
Meyer, M. C. (2006). Consistency and power in tests with shape-restricted alternatives. Journal of Statistical Planning and Inference , 136, 3931–3947.10.1016/j.jspi.2005.04.001CrossRefGoogle Scholar
Newey, W. K. (1990). Semiparametric efficiency bounds. Journal of Applied Econometrics , 5, 99–135.10.1002/jae.3950050202CrossRefGoogle Scholar
Newey, W. K. (1994). The asymptotic variance of semiparametric estimators. Econometrica , 62, 1349–1382.10.2307/2951752CrossRefGoogle Scholar
Newey, W. K., Hsieh, F. & Robins, J. M. (1998). Undersmoothing and bias corrected functional estimation. Working Paper 98-17, MIT.Google Scholar
Newey, W. K., Hsieh, F., & Robins, J. M. (2004). Twicing kernels and a small bias property of semiparametric estimators. Econometrica , 72, 947–962.10.1111/j.1468-0262.2004.00518.xCrossRefGoogle Scholar
Otsu, T., & Rai, Y. (2017). Bootstrap inference of matching estimators for average treatment effects. Journal of the American Statistical Association , 112, 1720–1732.10.1080/01621459.2016.1231613CrossRefGoogle Scholar
Qin, J., Yu, T., Li, P., Liu, H., & Chen, B. (2019). Using a monotone single-index model to stabilize the propensity score in missing data problems and causal inference. Statistics in Medicine , 38, 1442–1458.10.1002/sim.8048CrossRefGoogle ScholarPubMed
Rao, B. P. (1969) Estimation of a unimodal density. Sankhyā, A , 31, 23–36.Google Scholar
Rao, B. P. (1970). Estimation for distributions with monotone failure rate. Annals of Mathematical Statistics , 41, 507–519.10.1214/aoms/1177697091CrossRefGoogle Scholar
Ritov, J. M., & Ritov, Y. A. (1997). Toward a curse of dimensionality appropriate (CODA) asymptotic theory for semi-parametric models. Statistics in Medicine , 16, 285–319.Google Scholar
Robins, J., & Rotnitzky, A. (1995). Semiparametric efficiency in multivariate regression models with missing data. Journal of the American Statistical Association , 90, 122–129.10.1080/01621459.1995.10476494CrossRefGoogle Scholar
Robins, J. M., Rotnitzky, A., & Zhao, L. P. (1995). Analysis of semiparametric regression models for repeated outcomes in the presence of missing data. Journal of the American Statistical Association , 90, 106–121.10.1080/01621459.1995.10476493CrossRefGoogle Scholar
Robinson, P. M. (1988). Root-N-consistent semiparametric regression. Econometrica , 56, 931–954.10.2307/1912705CrossRefGoogle Scholar
Rosenbaum, P. R. (1989). Optimal matching for observational studies. Journal of the American Statistical Association , 84, 1024–1032.10.1080/01621459.1989.10478868CrossRefGoogle Scholar
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika , 70, 41–55.10.1093/biomet/70.1.41CrossRefGoogle Scholar
Rosenbaum, P. R., & Rubin, D. B. (1984). Reducing bias in observational studies using subclassification on the propensity score. Journal of the American Statistical Association , 79, 516–524.10.1080/01621459.1984.10478078CrossRefGoogle Scholar
Rosenbaum, P. R., & Rubin, D. B. (1985). Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. American Statistician , 39, 33–38.10.1080/00031305.1985.10479383CrossRefGoogle Scholar
Rothe, C. (2017). Robust confidence intervals for average treatment effects under limited overlap. Econometrica , 85, 645–660.10.3982/ECTA13141CrossRefGoogle Scholar
Rothe, C., & Firpo, S. (2019). Properties of doubly robust estimators when nuisance functions are estimated nonparametrically. Econometric Theory , 35, 1048–1087.10.1017/S0266466618000385CrossRefGoogle Scholar
Scharfstein, D. O., Rotnitzky, A., & Robins, J. M. (1999). Adjusting for nonignorable drop-out using semiparametric nonresponse models. Journal of the American Statistical Association , 94, 1096–1120.10.1080/01621459.1999.10473862CrossRefGoogle Scholar
Sherman, R. P. (1993). The limiting distribution of the maximum rank correlation estimator. Econometrica , 61, 123–137.10.2307/2951780CrossRefGoogle Scholar
van de Geer, S. (2000). Empirical processes in M-estimation . Cambridge University Press.Google Scholar
van der Vaart, A. (1991). On differentiable functionals. Annals of Statistics , 19, 178–204.Google Scholar
van der Vaart, A. W., & Wellner, J. A. (1996). Weak convergence and empirical processes . Springer.10.1007/978-1-4757-2545-2CrossRefGoogle Scholar
Wright, F. T. (1981). The asymptotic behavior of monotone regression estimates. Annals of Statistics , 9, 443–448.10.1214/aos/1176345411CrossRefGoogle Scholar
Xu, M. (2021). Essays in semiparametric estimation and inference with monotonicity constraints [Doctoral dissertation]. London School of Economics and Political Science.Google Scholar
Yu, K. (2014). On partial linear additive isotonic regression. Journal of the Korean Statistical Society , 43, 11–17.10.1016/j.jkss.2013.07.003CrossRefGoogle Scholar
Yuan, A., Yin, A., & Tan, M. T. (2021). Enhanced doubly robust procedure for causal inference. Statistics in Biosciences , 13, 454–478.10.1007/s12561-021-09300-yCrossRefGoogle Scholar
Supplementary material: File

Xu and Otsu supplementary material

Xu and Otsu supplementary material
Download Xu and Otsu supplementary material(File)
File 249.7 KB