Abstract:Consider estimation of average treatment effects with multi-valued treatments using augmented inverse probability weighted (IPW) estimators, depending on outcome regression and propensity score models in high-dimensional settings. These regression models are often fitted by regularized likelihood-based estimation, while ignoring how the fitted functions are used in the subsequent inference about the treatment parameters. Such separate estimation can be associated with known difficulties in existing methods. We develop regularized calibrated estimation for fitting propensity score and outcome regression models, where sparsity-including penalties are employed to facilitate variable selection but the loss functions are carefully chosen such that valid confidence intervals can be obtained under possible model misspecification. Unlike in the case of binary treatments, the usual augmented IPW estimator is generalized by allowing different copies of coefficient estimators in outcome regression to ensure just-identification. For propensity score estimation, the new loss function and estimating functions are directly tied to achieving covariate balance between weighted treatment groups. We develop practical numerical algorithms for computing the regularized calibrated estimators with group Lasso by innovatively exploiting Fisher scoring, and provide rigorous high-dimensional analysis for the resulting augmented IPW estimators under suitable sparsity conditions, while tackling technical issues absent or overlooked in previous analyses. We present simulation studies and an empirical application to estimate the effects of maternal smoking on birth weights. The proposed methods are implemented in the R package mRCAL.

Robust Inference Using Inverse Probability Weighting

Heavy Tail Robust Estimation and Inference for Average Treatment Effects

Reflection on modern methods: a note on variance estimation when using inverse probability weighting to handle attrition in cohort studies

Sharp Sensitivity Analysis for Inverse Propensity Weighting via Quantile Balancing

Smaller Confidence Intervals From IPW Estimators via Data-Dependent Coarsening

Inverse probability of treatment weighting with generalized linear outcome models for doubly robust estimation

Privacy-Preserving Causal Inference via Inverse Probability Weighting

Parameter Estimations on Zero-Inflated Negative Binomial Model with Incomplete Data

Estimating causal effects for binary outcomes using per-decision inverse probability weighting

Nonparametric augmented probability weighting with sparsity

A NEW PERSPECTIVE ON ROBUST M-ESTIMATION: FINITE SAMPLE THEORY AND APPLICATIONS TO DEPENDENCE-ADJUSTED MULTIPLE TESTING

Inverse Probability-Weighted Estimation for Dynamic Structural Equation Model with Missing Data

Doubly Robust Estimation of Local Average Treatment Effects Using Inverse Probability Weighted Regression Adjustment

Overlap, matching, or entropy weights: what are we weighting for?

A New Central Limit Theorem for the Augmented IPW Estimator: Variance Inflation, Cross-Fit Covariance and Beyond

A doubly robust estimator for the Mann Whitney Wilcoxon Rank Sum Test when applied for causal inference in observational studies

An Inverse Probability Weighted Regression Method that Accounts for Right-censoring for Causal Inference with Multiple Treatments and a Binary Outcome

Double robust semiparametric weighted M-estimators of a zero-inflated Poisson regression with missing data in covariates

On Flexible Inverse Probability of Treatment and Intensity Weighting: Informative Censoring, Variable Inclusion, and Weight Trimming

High-dimensional model-assisted inference for treatment effects with multi-valued treatments

On Frequency and Probability Weights: An In‐Depth Look at Duelling Weights