Minimal dispersion approximately balancing weights: asymptotic properties and practical considerations
From MaRDI portal
(Redirected from Publication:5222222)
Abstract: Weighting methods are widely used to adjust for covariates in observational studies, sample surveys, and regression settings. In this paper, we study a class of recently proposed weighting methods which find the weights of minimum dispersion that approximately balance the covariates. We call these weights "minimal weights" and study them under a common optimization framework. The key observation is the connection between approximate covariate balance and shrinkage estimation of the propensity score. This connection leads to both theoretical and practical developments. From a theoretical standpoint, we characterize the asymptotic properties of minimal weights and show that, under standard smoothness conditions on the propensity score function, minimal weights are consistent estimates of the true inverse probability weights. Also, we show that the resulting weighting estimator is consistent, asymptotically normal, and semiparametrically efficient. From a practical standpoint, we present a finite sample oracle inequality that bounds the loss incurred by balancing more functions of the covariates than strictly needed. This inequality shows that minimal weights implicitly bound the number of active covariate balance constraints. We finally provide a tuning algorithm for choosing the degree of approximate balance in minimal weights. We conclude the paper with four empirical studies that suggest approximate balance is preferable to exact balance, especially when there is limited overlap in covariate distributions. In these studies, we show that the root mean squared error of the weighting estimator can be reduced by as much as a half with approximate balance.
Recommendations
- Balancing covariates via propensity score weighting
- Stable weights that balance covariates for estimation with incomplete outcome data
- A framework for covariate balance using Bregman distances
- Kernel balancing: a flexible non-parametric weighting procedure for estimating causal effects
- Covariate association eliminating weights: a unified weighting framework for causal effect estimation
Cited in
(49)- Augmented minimax linear estimation
- Estimation of partially conditional average treatment effect by double kernel-covariate balancing
- Complex discontinuity designs using covariates: impact of school grade retention on later life outcomes in Chile
- Balancing covariates via propensity score weighting
- Building Representative Matched Samples With Multi-Valued Treatments in Large Observational Studies
- Stable weights that balance covariates for estimation with incomplete outcome data
- Comparison of covariate balance weighting methods in estimating treatment effects
- Evaluating and improving a matched comparison of antidepressants and bone density
- Joint calibrated estimation of inverse probability of treatment and censoring weights for marginal structural models
- Improving Trial Generalizability Using Observational Studies
- Calibration Techniques Encompassing Survey Sampling, Missing Data Analysis and Causal Inference
- Large Sample Properties of Matching for Balance
- Hospital quality risk standardization via approximate balancing weights
- Varying impacts of letters of recommendation on college admissions
- Balancing weights for region-level analysis: the effect of medicaid expansion on the uninsurance rate among states that did not expand medicaid
- A note on semiparametric efficient generalization of causal effects from randomized trials to target populations
- Transfer Learning of Individualized Treatment Rules from Experimental to Real-World Data
- Projected state-action balancing weights for offline reinforcement learning
- RKHS-based covariate balancing for survival causal effect estimation
- Causal effect estimation for multivariate continuous treatments
- Entropy balancing for causal generalization with target sample summary information
- Approximate balancing weights for clustered observational study designs
- Balancing versus modelling in weighted analysis of non-randomised studies with survival outcomes: a simulation study
- Balancing vs modeling approaches to weighting in practice
- More robust estimation of average treatment effects using kernel optimal matching in an observational study of spine surgical interventions
- Transporting experimental results with entropy balancing
- A calibration approach to transportability and data-fusion with observational data
- Mahalanobis balancing: a multivariate perspective on approximate covariate balancing
- A framework for causal estimand selection under positivity violations
- Confidence interval construction for causally generalized estimates with target sample summary information
- Covariate balancing with measurement error
- Generalized Bayesian inference for causal effects using the covariate balancing procedure
- Weighted Euclidean balancing for a matrix exposure in estimating causal effect
- Federated Adaptive Causal Estimation (FACE) of Target Treatment Effects
- Variance-based sensitivity analysis for weighting estimators results in more informative bounds
- Individualized treatment rules based on adaptive transfer-dragonnet
- Statistical balancing as an unconstrained optimisation problem
- One-Step Weighting to Generalize and Transport Treatment Effect Estimates to a Target Population
- A calibrated sensitivity analysis for weighted causal decompositions
- So many choices: a guide to selecting among methods to adjust for observed confounders
- Optimization for Calibration of Survey Weights under a Large Number of Conflicting Constraints
- Sensitivity analysis for the generalization of experimental results
- Dynamic covariate balancing: estimating treatment effects over time with potential local projections
- Variance estimation for ridge calibration estimation in survey sampling
- Augmented balancing weights as linear regression
- Discussion of: ``Augmented balancing weights as linear regression
- Discussion of: ``Augmented balancing weights as linear regression
- A review of off-policy evaluation in reinforcement learning
- Targeted quality measurement of health care providers
This page was built for publication: Minimal dispersion approximately balancing weights: asymptotic properties and practical considerations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5222222)
