pith. sign in

arxiv: 2605.05399 · v1 · submitted 2026-05-06 · 📊 stat.ME

Causal Effect Estimation on Restricted Mean Survival Time in Case-Cohort Studies via a Matching Design

Andy Ni , Wei-En Lu , Bo Lu This is my paper

Pith reviewed 2026-05-08 15:59 UTC · model grok-4.3

classification 📊 stat.ME
keywords case-cohort designrestricted mean survival timecausal inferencetemplate matchingsurvival analysisconfounder adjustmentpropensity score
0
0 comments X

The pith

Template matching estimates the causal difference in restricted mean survival time under stratified case-cohort sampling.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The authors develop a method to estimate the marginal causal effect of an exposure on the difference in restricted mean survival times when data follow a stratified case-cohort design. They adjust for measured confounders by applying a template matching procedure that pairs individuals without requiring equal numbers of exposed and unexposed subjects. A sympathetic reader would care because large observational studies often use case-cohort sampling to limit the cost of measuring covariates on everyone, yet still need valid causal estimates for time-to-event outcomes where the hazard ratio may not be the best summary. The paper derives asymptotic properties of the resulting estimators, supplies a bootstrap variance procedure, and checks performance with simulations that compare template matching to conventional matching and to matching on the propensity score. It closes with an application to the ARIC cohort that estimates the effect of high-sensitivity C-reactive protein on coronary heart disease-free survival.

Core claim

Under the stratified case-cohort design, a template matching procedure can be used to estimate the difference in restricted mean survival times between exposed and unexposed groups while adjusting for confounders; the estimators are asymptotically normal and their variances can be obtained by bootstrap.

What carries the argument

Template matching design that permits flexible sample sizes for the exposed and unexposed groups while balancing measured covariates or propensity scores.

Load-bearing premise

Matching on observed covariates or propensity scores balances all relevant confounders, with no unmeasured confounding remaining after matching.

What would settle it

A dataset or simulation containing known unmeasured confounding in which the template-matched RMST estimator fails to recover the true causal difference would falsify the central claim.

Figures

Figures reproduced from arXiv: 2605.05399 by Andy Ni, Bo Lu, Wei-En Lu.

Figure 1
Figure 1. Figure 1: Standardized mean differences of covariates used in the ARIC study analysis before and after matching. view at source ↗
read the original abstract

In large observational studies, the case-cohort design is commonly used to reduce the cost associated with covariate measurement. For survival outcomes, literature has suggested that the restricted mean survival time (RMST) be a more appropriate marginal causal effect measure than the hazard ratio. In this paper, we develop a marginal causal effect estimation method for RMST difference under the stratified case-cohort design. We adjust for measured confounders using an innovative template matching design. Compared with conventional matching designs, template matching allows greater flexibility in the sample sizes of the exposed and unexposed groups. We establish the asymptotic properties of the proposed causal effect estimators and develop a bootstrap procedure to estimate their variances. By conducting comprehensive simulation studies, we evaluate the finite sample performance of the proposed estimators, demonstrate the advantage of template matching over conventional matching, and compare between matching on propensity score and matching on covariates. Finally, we apply the proposed methods to the Atherosclerosis Risk in Communities (ARIC) Study to estimate the marginal causal effect of serum hs-CRP level on the coronary heart disease-free survival.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper proposes a template matching design to estimate the marginal causal effect on restricted mean survival time (RMST) difference in stratified case-cohort studies. It adjusts for measured confounders by matching on covariates or propensity scores, derives asymptotic properties of the resulting estimators, introduces a bootstrap variance estimator, evaluates finite-sample performance via simulations (including comparisons to conventional matching), and applies the method to the ARIC study to assess the effect of hs-CRP on coronary heart disease-free survival.

Significance. If the central consistency result holds after incorporating the case-cohort sampling design, the work would provide a flexible alternative to standard matching for RMST-based causal inference in cost-efficient sampling schemes, with potential advantages in balancing exposed and unexposed groups. The simulation studies and bootstrap procedure offer concrete empirical and practical support.

major comments (3)
  1. [§3] §3 (Template Matching Design and Estimator): The description of the matching procedure and subsequent RMST difference estimator does not specify how the known stratum-specific sampling probabilities (or inverse-probability weights) from the stratified case-cohort design are incorporated. Without explicit design-based weighting at the matching or estimation stage, the matched subsample's covariate and survival distributions remain distorted relative to the full cohort, so the estimator is not guaranteed to be consistent for the population marginal causal effect.
  2. [§4] §4 (Asymptotic Properties): The consistency and asymptotic normality claims appear to be derived under an implicit simple-random-sampling assumption rather than under the actual case-cohort sampling mechanism. Because the central claim is that the procedure identifies the marginal RMST contrast in the target population, this omission is load-bearing; a revised derivation that explicitly conditions on the sampling indicators and weights is required.
  3. [§5] §5 (Bootstrap Procedure): The proposed bootstrap is described as resampling from the matched sample, but it is unclear whether it resamples the original case-cohort sampling process or accounts for the variability induced by the template matching step itself. This affects the validity of the variance estimator for inference on the population parameter.
minor comments (2)
  1. [§2] Notation for the RMST functional and the template-matching distance metric should be introduced earlier and used consistently across sections to improve readability.
  2. [§6] Simulation tables would benefit from explicit reporting of the effective sample sizes after matching and the realized balance metrics (e.g., standardized mean differences) under each design.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive review of our manuscript. The comments correctly identify areas where the integration of the stratified case-cohort sampling design required more explicit description. We have revised the manuscript to address each point by clarifying the use of sampling weights, updating the asymptotic derivations, and refining the bootstrap procedure.

read point-by-point responses

  1. Referee: [§3] §3 (Template Matching Design and Estimator): The description of the matching procedure and subsequent RMST difference estimator does not specify how the known stratum-specific sampling probabilities (or inverse-probability weights) from the stratified case-cohort design are incorporated. Without explicit design-based weighting at the matching or estimation stage, the matched subsample's covariate and survival distributions remain distorted relative to the full cohort, so the estimator is not guaranteed to be consistent for the population marginal causal effect.

    Authors: We agree that the original description in Section 3 lacked sufficient explicitness regarding the stratum-specific sampling probabilities. We have revised this section to state that template matching is performed within each stratum using the known sampling probabilities, and that the RMST difference estimator is constructed with inverse-probability-of-sampling weights applied to the matched observations. These weights ensure the estimator remains consistent for the population marginal causal effect under the stratified case-cohort design. revision: yes

  2. Referee: [§4] §4 (Asymptotic Properties): The consistency and asymptotic normality claims appear to be derived under an implicit simple-random-sampling assumption rather than under the actual case-cohort sampling mechanism. Because the central claim is that the procedure identifies the marginal RMST contrast in the target population, this omission is load-bearing; a revised derivation that explicitly conditions on the sampling indicators and weights is required.

    Authors: We acknowledge that the asymptotic results in Section 4 were not derived with explicit conditioning on the case-cohort sampling indicators. We have revised the proofs to incorporate the sampling mechanism directly: the consistency and asymptotic normality are now established by conditioning on the stratum-specific sampling indicators and including the inverse-probability weights in the influence function and variance expressions. The revised derivation confirms consistency for the population marginal RMST contrast. revision: yes

  3. Referee: [§5] §5 (Bootstrap Procedure): The proposed bootstrap is described as resampling from the matched sample, but it is unclear whether it resamples the original case-cohort sampling process or accounts for the variability induced by the template matching step itself. This affects the validity of the variance estimator for inference on the population parameter.

    Authors: We thank the referee for noting the ambiguity in the bootstrap description. We have revised Section 5 to specify that the bootstrap procedure resamples the original stratified case-cohort sampling process (including the sampling indicators within strata) and then reapplies the template matching within each replicate. This accounts for variability from both the sampling design and the matching step, yielding valid variance estimates for the population parameter. revision: yes

Circularity Check

0 steps flagged

No circularity: derivation uses standard causal assumptions plus new matching design with independent asymptotic justification.

full rationale

The paper proposes a template matching approach for RMST causal estimation under stratified case-cohort sampling, derives asymptotic properties for the estimators, and proposes a bootstrap for variance. No step reduces a claimed prediction or result to a fitted quantity by construction, nor does any load-bearing premise collapse to a self-citation chain or self-definition. The central identification relies on the matching balancing observed confounders under the design, with simulations and real-data application serving as external checks rather than tautological re-derivations. This is the common honest case of a self-contained methodological contribution.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Central claim rests on standard observational causal assumptions plus validity of the template matching procedure under case-cohort sampling. No free parameters or invented entities are identifiable from the abstract.

axioms (2)
  • domain assumption No unmeasured confounding
    Required for causal interpretation of matching-based estimators in observational data.
  • domain assumption Positivity / overlap for matching
    Template matching requires sufficient overlap between exposed and unexposed groups.

pith-pipeline@v0.9.0 · 5485 in / 1137 out tokens · 40312 ms · 2026-05-08T15:59:48.734376+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages

  1. [1]

    Reducing bias in observational studies using subclassification on the propensity score.Journal of the American statistical Association, 79(387):516–524, 1984

    Paul R Rosenbaum and Donald B Rubin. Reducing bias in observational studies using subclassification on the propensity score.Journal of the American statistical Association, 79(387):516–524, 1984

    work page 1984

  2. [2]

    Performance of tests of significance based on stratification by a multivariate confounder score or by a propensity score.Journal of clinical epidemiology, 42(4):317–324, 1989

    E Francis Cook and Lee Goldman. Performance of tests of significance based on stratification by a multivariate confounder score or by a propensity score.Journal of clinical epidemiology, 42(4):317–324, 1989

    work page 1989

  3. [3]

    Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study.Statistics in medicine, 23(19):2937–2960, 2004

    Jared K Lunceford and Marie Davidian. Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study.Statistics in medicine, 23(19):2937–2960, 2004

    work page 2004

  4. [4]

    An introduction to propensity score methods for reducing the effects of confounding in observa- tional studies.Multivariate behavioral research, 46(3):399–424, 2011

    Peter C Austin. An introduction to propensity score methods for reducing the effects of confounding in observa- tional studies.Multivariate behavioral research, 46(3):399–424, 2011

    work page 2011

  5. [5]

    Causal effect estimation on restricted mean survival time under case-cohort design via propensity score stratification.Lifetime Data Analysis, pages 1–34, 2025

    Wei-En Lu and Ai Ni. Causal effect estimation on restricted mean survival time under case-cohort design via propensity score stratification.Lifetime Data Analysis, pages 1–34, 2025

    work page 2025

  6. [6]

    The central role of the propensity score in observational studies for causal effects.Biometrika, 70(1):41–55, 1983

    Paul R Rosenbaum and Donald B Rubin. The central role of the propensity score in observational studies for causal effects.Biometrika, 70(1):41–55, 1983

    work page 1983

  7. [7]

    Springer, 2010

    Paul R Rosenbaum, P Rosenbaum, and Briskman.Design of observational studies, volume 10. Springer, 2010

    work page 2010

  8. [8]

    Matching methods for causal inference: A review and a look forward.Statistical science: a review journal of the Institute of Mathematical Statistics, 25(1):1, 2010

    Elizabeth A Stuart. Matching methods for causal inference: A review and a look forward.Statistical science: a review journal of the Institute of Mathematical Statistics, 25(1):1, 2010

    work page 2010

  9. [9]

    Flexible template matching for observational study design.Statistics in Medicine, 42(11):1760–1778, 2023

    Ruochen Zhao and Bo Lu. Flexible template matching for observational study design.Statistics in Medicine, 42(11):1760–1778, 2023

    work page 2023

  10. [10]

    On the role of the propensity score in efficient semiparametric estimation of average treatment effects.Econometrica, pages 315–331, 1998

    Jinyong Hahn. On the role of the propensity score in efficient semiparametric estimation of average treatment effects.Econometrica, pages 315–331, 1998

    work page 1998

  11. [11]

    Keisuke Hirano and Guido W Imbens. Estimation of causal effects using propensity score weighting: An application to data on right heart catheterization.Health Services and Outcomes research methodology, 2:259–278, 2001

    work page 2001

  12. [12]

    Efficient estimation of average treatment effects using the estimated propensity score.Econometrica, 71(4):1161–1189, 2003

    Keisuke Hirano, Guido W Imbens, and Geert Ridder. Efficient estimation of average treatment effects using the estimated propensity score.Econometrica, 71(4):1161–1189, 2003

    work page 2003

  13. [13]

    Nonparametric estimation of average treatment effects under exogeneity: A review.Review of Economics and statistics, 86(1):4–29, 2004

    Guido W Imbens. Nonparametric estimation of average treatment effects under exogeneity: A review.Review of Economics and statistics, 86(1):4–29, 2004

    work page 2004

  14. [14]

    A structural approach to selection bias

    Miguel A Hernán, Sonia Hernández-Díaz, and James M Robins. A structural approach to selection bias. Epidemiology, 15(5):615–625, 2004

    work page 2004

  15. [15]

    The hazards of hazard ratios.Epidemiology, 21(1):13–15, 2010

    Miguel A Hernán. The hazards of hazard ratios.Epidemiology, 21(1):13–15, 2010

    work page 2010

  16. [16]

    Does cox analysis of a randomized survival study yield a causal treatment effect?Lifetime data analysis, 21(4):579–593, 2015

    Odd O Aalen, Richard J Cook, and Kjetil Røysland. Does cox analysis of a randomized survival study yield a causal treatment effect?Lifetime data analysis, 21(4):579–593, 2015

    work page 2015

  17. [17]

    A note on the noncollapsibility of rate differences and rate ratios.Epidemiology, 27(3):356–359, 2016

    Arvid Sjölander, Elisabeth Dahlqwist, and Johan Zetterqvist. A note on the noncollapsibility of rate differences and rate ratios.Epidemiology, 27(3):356–359, 2016

    work page 2016

  18. [18]

    Marginal structural models and causal inference in epidemiology, 2000

    James M Robins, Miguel Angel Hernan, and Babette Brumback. Marginal structural models and causal inference in epidemiology, 2000

    work page 2000

  19. [19]

    JO Irwin. The standard error of an estimate of expectation of life, with special reference to expectation of tumourless life in experiments with mice.Epidemiology & Infection, 47(2):188–189, 1949

    work page 1949

  20. [20]

    Predicting the restricted mean event time with the subject’s baseline covariates in survival analysis.Biostatistics, 15(2):222–233, 2014

    Lu Tian, Lihui Zhao, and Lee-Jen Wei. Predicting the restricted mean event time with the subject’s baseline covariates in survival analysis.Biostatistics, 15(2):222–233, 2014. 12 APREPRINT- MAY8, 2026

    work page 2014

  21. [21]

    On the restricted mean survival time curve in survival analysis.Biometrics, 72(1):215–221, 2016

    Lihui Zhao, Brian Claggett, Lu Tian, Hajime Uno, Marc A Pfeffer, Scott D Solomon, Lorenzo Trippa, and Lee-Jen Wei. On the restricted mean survival time curve in survival analysis.Biometrics, 72(1):215–221, 2016

    work page 2016

  22. [22]

    Modeling restricted mean survival time under general censoring mechanisms

    Xin Wang and Douglas E Schaubel. Modeling restricted mean survival time under general censoring mechanisms. Lifetime data analysis, 24(1):176–199, 2018

    work page 2018

  23. [23]

    Double-robust semiparametric estimator for differences in restricted mean lifetimes in observational studies.Biometrics, 68(4):999–1009, 2012

    Min Zhang and Douglas E Schaubel. Double-robust semiparametric estimator for differences in restricted mean lifetimes in observational studies.Biometrics, 68(4):999–1009, 2012

    work page 2012

  24. [24]

    Adjusted restricted mean survival times in observational studies.Statistics in medicine, 38(20):3832–3860, 2019

    Sarah C Conner, Lisa M Sullivan, Emelia J Benjamin, Michael P LaValley, Sandro Galea, and Ludovic Trinquart. Adjusted restricted mean survival times in observational studies.Statistics in medicine, 38(20):3832–3860, 2019

    work page 2019

  25. [25]

    Stratified restricted mean survival time model for marginal causal effect in observational survival data.Annals of epidemiology, 64:149–154, 2021

    Ai Ni, Zihan Lin, and Bo Lu. Stratified restricted mean survival time model for marginal causal effect in observational survival data.Annals of epidemiology, 64:149–154, 2021

    work page 2021

  26. [26]

    Matched design for marginal causal effect on restricted mean survival time in observational studies.Journal of Causal Inference, 11(1):20220035, 2023

    Zihan Lin, Ai Ni, and Bo Lu. Matched design for marginal causal effect on restricted mean survival time in observational studies.Journal of Causal Inference, 11(1):20220035, 2023

    work page 2023

  27. [27]

    A case-cohort design for epidemiologic cohort studies and disease prevention trials.Biometrika, 73(1):1–11, 1986

    Ross L Prentice. A case-cohort design for epidemiologic cohort studies and disease prevention trials.Biometrika, 73(1):1–11, 1986

    work page 1986

  28. [28]

    Exposure stratified case-cohort designs.Lifetime data analysis, 6(1):39–58, 2000

    Ornulf Borgan, Bryan Langholz, Sven Ove Samuelsen, Larry Goldstein, and Janice Pogoda. Exposure stratified case-cohort designs.Lifetime data analysis, 6(1):39–58, 2000

    work page 2000

  29. [29]

    Asymptotic distribution theory and efficiency results for case-cohort studies

    Steven G Self and Ross L Prentice. Asymptotic distribution theory and efficiency results for case-cohort studies. The Annals of Statistics, pages 64–81, 1988

    work page 1988

  30. [30]

    Robust variance estimation for the case-cohort design.Biometrics, pages 1064–1072, 1994

    William E Barlow. Robust variance estimation for the case-cohort design.Biometrics, pages 1064–1072, 1994

    work page 1994

  31. [31]

    Likelihood analysis of multi-state models for disease incidence and mortality

    JD Kalbfleisch and JF Lawless. Likelihood analysis of multi-state models for disease incidence and mortality. Statistics in medicine, 7(1-2):149–160, 1988

    work page 1988

  32. [32]

    Improving the efficiency of relative-risk estimation in case-cohort studies.Journal of the American Statistical Association, 99(467):832–844, 2004

    Michal Kulich and DY Lin. Improving the efficiency of relative-risk estimation in case-cohort studies.Journal of the American Statistical Association, 99(467):832–844, 2004

    work page 2004

  33. [33]

    Multiple imputation analysis of case–cohort studies.Statistics in medicine, 30(13):1595–1607, 2011

    Helena Marti and Michel Chavance. Multiple imputation analysis of case–cohort studies.Statistics in medicine, 30(13):1595–1607, 2011

    work page 2011

  34. [34]

    Invited commentary: propensity scores.American journal of epidemiology, 150(4):327–333, 1999

    Marshall M Joffe and Paul R Rosenbaum. Invited commentary: propensity scores.American journal of epidemiology, 150(4):327–333, 1999

    work page 1999

  35. [35]

    On the estimation and use of propensity scores in case-control and case-cohort studies.American journal of epidemiology, 166(3):332–339, 2007

    Roger Månsson, Marshall M Joffe, Wenguang Sun, and Sean Hennessy. On the estimation and use of propensity scores in case-control and case-cohort studies.American journal of epidemiology, 166(3):332–339, 2007

    work page 2007

  36. [36]

    Causal inference in outcome-dependent two-phase sampling designs.Journal of the Royal Statistical Society Series B: Statistical Methodology, 71(5):947– 969, 2009

    Weiwei Wang, Daniel Scharfstein, Zhiqiang Tan, and Ellen J MacKenzie. Causal inference in outcome-dependent two-phase sampling designs.Journal of the Royal Statistical Society Series B: Statistical Methodology, 71(5):947– 969, 2009

    work page 2009

  37. [37]

    Stephen R Cole, Michael G Hudgens, Phyllis C Tien, Kathryn Anastos, Lawrence Kingsley, Joan S Chmiel, and Lisa P Jacobson. Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident aids or death.American journal of epidemiology, 175(5):381–390, 2012

    work page 2012

  38. [38]

    Marginal structural cox models with case-cohort sampling.Statistica Sinica, 26(2):509, 2016

    Hana Lee, Michael G Hudgens, Jianwen Cai, and Stephen R Cole. Marginal structural cox models with case-cohort sampling.Statistica Sinica, 26(2):509, 2016

    work page 2016

  39. [39]

    Template matching for auditing hospital cost and quality.Health services research, 49(5):1446–1474, 2014

    Jeffrey H Silber, Paul R Rosenbaum, Richard N Ross, Justin M Ludwig, Wei Wang, Bijan A Niknam, Nabanita Mukherjee, Philip A Saynisch, Orit Even-Shoshan, Rachel R Kelz, et al. Template matching for auditing hospital cost and quality.Health services research, 49(5):1446–1474, 2014

    work page 2014

  40. [40]

    The atherosclerosis risk in communit (aric) study: design and objectives.American journal of epidemiology, 129(4):687–702, 1989

    Aric Investigators. The atherosclerosis risk in communit (aric) study: design and objectives.American journal of epidemiology, 129(4):687–702, 1989

    work page 1989

  41. [41]

    Estimating causal effects of treatments in randomized and nonrandomized studies.Journal of educational Psychology, 66(5):688, 1974

    Donald B Rubin. Estimating causal effects of treatments in randomized and nonrandomized studies.Journal of educational Psychology, 66(5):688, 1974

    work page 1974

  42. [42]

    Large sample properties of matching estimators for average treatment effects.econometrica, 74(1):235–267, 2006

    Alberto Abadie and Guido W Imbens. Large sample properties of matching estimators for average treatment effects.econometrica, 74(1):235–267, 2006

    work page 2006

  43. [43]

    Bias-corrected matching estimators for average treatment effects.Journal of Business & Economic Statistics, 29(1):1–11, 2011

    Alberto Abadie and Guido W Imbens. Bias-corrected matching estimators for average treatment effects.Journal of Business & Economic Statistics, 29(1):1–11, 2011

    work page 2011

  44. [44]

    On the inconsistency of matching without replacement.Biometrika, 109(2):551–558, 2022

    Fredrik Sävje. On the inconsistency of matching without replacement.Biometrika, 109(2):551–558, 2022. 13 APREPRINT- MAY8, 2026

    work page 2022

  45. [45]

    Optimal matching for observational studies.Journal of the American Statistical Association, 84(408):1024–1032, 1989

    Paul R Rosenbaum. Optimal matching for observational studies.Journal of the American Statistical Association, 84(408):1024–1032, 1989

    work page 1989

  46. [46]

    Matching design for augmenting the control arm of a randomized controlled trial using real-world data.Journal of Biopharmaceutical Statistics, 32(1):124–140, 2022

    Yingying Liu, Bo Lu, Richard Foster, Yiwei Zhang, Z John Zhong, Ming-Hui Chen, and Peng Sun. Matching design for augmenting the control arm of a randomized controlled trial using real-world data.Journal of Biopharmaceutical Statistics, 32(1):124–140, 2022

    work page 2022

  47. [47]

    Theory and applications of hazard plotting for censored failure data.Technometrics, 14(4):945–966, 1972

    Wayne Nelson. Theory and applications of hazard plotting for censored failure data.Technometrics, 14(4):945–966, 1972

    work page 1972

  48. [48]

    A martingale representation for matching estimators.Journal of the American Statistical Association, 107(498):833–843, 2012

    Alberto Abadie and Guido W Imbens. A martingale representation for matching estimators.Journal of the American Statistical Association, 107(498):833–843, 2012

    work page 2012

  49. [49]

    On the Generalised Distance in Statistics

    Prasanta Chandra Mahalanobis. Reprint of: Mahalanobis, P.C. (1936) "On the Generalised Distance in Statistics.". Sankhya A, 80(1):1–7, December 1936

    work page 1936

  50. [50]

    On the failure of the bootstrap for matching estimators.Econometrica, 76(6):1537–1557, 2008

    Alberto Abadie and Guido W Imbens. On the failure of the bootstrap for matching estimators.Econometrica, 76(6):1537–1557, 2008

    work page 2008

  51. [51]

    Nonparametric bootstrap for propensity score matching estimators.Statistics & Probability Letters, 208:110069, 2024

    Hugo Bodory, Lorenzo Camponovo, Martin Huber, and Michael Lechner. Nonparametric bootstrap for propensity score matching estimators.Statistics & Probability Letters, 208:110069, 2024

    work page 2024

  52. [52]

    The use of bootstrapping when using propensity-score matching without replacement: a simulation study.Statistics in medicine, 33(24):4306–4319, 2014

    Peter C Austin and Dylan S Small. The use of bootstrapping when using propensity-score matching without replacement: a simulation study.Statistics in medicine, 33(24):4306–4319, 2014

    work page 2014

  53. [53]

    Robust post-matching inference.Journal of the American Statistical Association, 117(538):983–995, 2022

    Alberto Abadie and Jann Spiess. Robust post-matching inference.Journal of the American Statistical Association, 117(538):983–995, 2022

    work page 2022

  54. [54]

    Christie M Ballantyne, Ron C Hoogeveen, Heejung Bang, Josef Coresh, Aaron R Folsom, Gerardo Heiss, and A Richey Sharrett. Lipoprotein-associated phospholipase a2, high-sensitivity c-reactive protein, and risk for incident coronary heart disease in middle-aged men and women in the atherosclerosis risk in communities (aric) study.Circulation, 109(7):837–842, 2004

    work page 2004

  55. [55]

    Combining propensity score matching with additional adjustments for prognostic covariates.Journal of the American Statistical Association, 95(450):573–585, 2000

    Donald B Rubin and Neal Thomas. Combining propensity score matching with additional adjustments for prognostic covariates.Journal of the American Statistical Association, 95(450):573–585, 2000

    work page 2000

  56. [56]

    Springer Science & Business Media, 2013

    Aad Vaart and Jon Wellner.Weak convergence and empirical processes: with applications to statistics. Springer Science & Business Media, 2013

    work page 2013

  57. [57]

    Springer Science & Business Media, 2012

    Per K Andersen, Ornulf Borgan, Richard D Gill, and Niels Keiding.Statistical models based on counting processes. Springer Science & Business Media, 2012. A Proof of Lemma 1 We first establish the asymptotic distribution of√m[ ˆH 1 1(t)−H 1 1(t)]. √m[ ˆH 1 1(t)−H 1 1(t)] = √m Z t 0 Pm i=1 dN1i(u)Pm i=1 Y1i(u)ρ1i −H 1 1(t) = Z t 0 mPm i=1 Y1i(u)ρ1i 1√m mX i...

    work page 2012