aipw estimation

To address the dual limitations of the Augmented Inverse Probability Weighting (AIPW) estimator—instability under extreme propensity scores (near 0 or 1) and the Propensity Score Matching (PSM) estimator—dependence on correct model specification and lack of semiparametric efficiency—this paper proposes the Augmented Matching Weighting (AMW) estimator. AMW innovatively replaces inverse-propensity weights with adaptive matching weights based on a non-fixed number of matches $K$, and introduces a cross-validation procedure guided by unbiasedness to select $K$. Theoretically, AMW achieves double robustness and semiparametric efficiency, and supports valid inference via the naïve bootstrap. Simulation and empirical studies demonstrate that AMW significantly outperforms both AIPW and PSM in stability under extreme propensity scores, yields more accurate variance estimation, and maintains computational simplicity.

This paper addresses model misspecification bias in health technology assessment when randomized controlled trials are infeasible and treatment networks are disconnected—specifically, in external-control single-arm trials and unanchored indirect comparisons. We propose a doubly robust augmented Matching-Adjusted Indirect Comparison (MAIC) estimator that integrates conditional outcome modeling with entropy balancing weights, overcoming the single-robustness limitation of conventional MAIC. In simulations, our method matches the performance of G-computation and substantially outperforms non-augmented weighting approaches. Feasibility is further demonstrated under realistic scenarios with missing individual-level data. Our key contributions are: (i) the first doubly robust implementation within the MAIC framework; (ii) unified handling of both external-control and unanchored indirect comparisons; and (iii) significantly enhanced robustness and reliability of treatment effect estimation.

This paper investigates the impact of covariate balancing on average treatment effect (ATE) and average treatment effect on the treated (ATT) estimation, and establishes the theoretical foundation for numerical equivalence among inverse probability weighting (IPW), augmented IPW (AIPW), and IPW-regression adjustment (IPWRA) estimators. We rigorously prove that when propensity scores are estimated via covariate balancing methods—such as inverse probability tilting (IPT) for ATE or covariate-balancing propensity score (CBPS) for ATT—the three estimators are algebraically identical, and their weights are automatically normalized. This equivalence unifies the theoretical frameworks of weighted and doubly robust estimation, substantially improving finite-sample stability and precision. Moreover, the result enables a novel analytical pathway for identifying the local average treatment effect (LATE) under unmeasured confounding, thereby advancing model robustness and computational consistency in causal inference.

In pharmacoepidemiology, electronic health records frequently exhibit high missingness (>50%) in critical confounders and rare outcomes, leading to substantial bias in causal effect estimation. This study is the first to systematically compare doubly robust estimators—generalized raking and inverse-probability-weighted targeted maximum likelihood estimation (IPW-TMLE)—against conventional approaches—multiple imputation and inverse-probability weighting—within a plasmode simulation framework that preserves real-world data structure. Results demonstrate that both doubly robust estimators markedly reduce bias and mean squared error; IPW-TMLE consistently achieves superior performance across most scenarios. Furthermore, the study proposes a practical, bias–variance trade-off–informed guideline for selecting analytical methods. This work establishes an empirical benchmark and provides actionable recommendations for confounding control in observational studies characterized by high missingness and low event rates.

This paper addresses unbiased estimation of the marginal counterfactual cumulative incidence function (CIF) in observational studies under competing risks. We propose a modified Nelson–Aalen estimator based on inverse probability weighting (IPW), incorporating estimated propensity scores. Our key contribution is the first rigorous asymptotic theory for such an adjusted estimator within the competing risks framework: we explicitly characterize the additional variability induced by propensity score estimation and prove its asymptotically negligible impact on final inference. Leveraging counting process theory and influence function analysis, we derive influence functions for both the counterfactual cumulative hazard and the CIF, establishing consistency and asymptotic normality of the estimator. Simulation studies and real-data analyses demonstrate low bias, robust performance across scenarios, and accurate standard error estimation. The proposed method thus provides a theoretically sound and practically viable tool for causal inference in competing risks settings.

This study addresses the bias arising from negative weights in staggered difference-in-differences (DiD) designs, which distorts the average of heterogeneous treatment effects—particularly when time-varying covariates are present. Within a model-agnostic framework, the paper nonparametrically defines group-time, group, period, and dynamic average treatment effects for the first time. To this end, the authors propose an augmented inverse variance-weighted (AIVW) estimator that integrates the augmented inverse probability weighting (AIPW) principle, achieving double robustness: consistent estimation of target parameters is maintained even if either the outcome or propensity score model is misspecified. The asymptotic variance is derived via influence functions, enabling valid causal inference with time-varying covariates. Simulations and an empirical application to China’s college admissions policy demonstrate that the estimator performs reliably in finite samples and effectively identifies the causal effects of parallel志愿 and immediate enrollment policies.

This work addresses the challenge of mixed distributional shift, which simultaneously involves systematic bias and random perturbations—a setting where conventional reweighting methods struggle to balance bias correction with uncertainty quantification. The authors propose a novel framework that explicitly disentangles these two sources of shift: systematic bias is corrected via Augmented Inverse Distance Weighting (AIDW) and Augmented Inverse Hybrid Weighting (AIHW), while residual random perturbations are modeled as distributional uncertainty. Robust inference is achieved through variance-optimal data pooling. The method incorporates distributional distance to govern the bias–variance trade-off, offering both asymptotic theoretical guarantees and practical guidance for hyperparameter tuning. Evaluated on three real-world multicenter datasets, the approach substantially reduces mean squared error and improves empirical coverage, demonstrating particular robustness in scenarios where covariate shift correction typically underestimates uncertainty.

This study addresses the instability of policy value estimation in causal inference under practical positivity violations (limited overlap). The authors propose an Adaptive Targeted Maximum Likelihood Estimation (A-TMLE) framework that constructs a projected policy value parameter using a data-driven conditional average treatment effect (CATE) model and incorporates regularized TMLE to avoid direct reliance on inverse probability weighting. By integrating adaptive function approximation, projection techniques, and influence function theory, A-TMLE substantially enhances estimation robustness under limited overlap. Empirical evaluations on both simulated data and real-world right heart catheterization data demonstrate that A-TMLE achieves lower mean squared error, higher confidence interval coverage, and more compact and stable inference compared to standard IPW, AIPW, and conventional TMLE approaches.

This study addresses the weakened causal effect identification and heightened sensitivity to model misspecification arising from partially missing confounders in observational studies. It proposes the first doubly robust weighted least squares estimator (MI-WOLS), which integrates a propensity score model into the weighting structure of outcome regression within a missingness indicator framework. The resulting weights achieve covariate balance and ensure consistent estimation as long as either the propensity score model or the outcome model is correctly specified. Coupled with sandwich variance estimation, simulation studies demonstrate that MI-WOLS yields negligible bias, accurate variance estimates, and confidence intervals attaining nominal coverage. An application to kidney function data further confirms the method’s practical utility and interpretability.

This study addresses the instability of conventional causal estimation methods—such as augmented inverse probability weighting (AIPW) and targeted maximum likelihood estimation (TMLE)—under high-dimensional confounding and violations of the positivity assumption. The authors propose a residual-on-residual regression strategy: after separately fitting exposure and outcome models adjusted for confounders, they regress the resulting residuals using ordinary least squares to obtain a robust estimate of the causal effect in a partially linear model. This approach is computationally straightforward, highly interpretable, and maintains unbiasedness with well-calibrated confidence interval coverage even when positivity is violated. In simulations, it substantially outperforms existing methods and was successfully applied to the nuMoM2b dataset, revealing a modest negative association between higher vegetable intake density and preeclampsia risk.