What question did this study set out to answer?

This research aims to develop methods for estimating average causal effects when data on exposures and confounders are incomplete.

February 22, 2026Journal of Causal InferenceOpen Access

Estimating average causal effects with incomplete exposure and confounders

View Full Paper

Ask AI

Bookmark

This research aims to develop methods for estimating average causal effects when data on exposures and confounders are incomplete.
Proposed methods for estimating average causal effects with missing confounders.
Consideration of missingness at random (MAR) and missing not at random (MNAR) scenarios.
Development of doubly robust targeted maximum likelihood estimators (TMLE).
Simulation studies to compare standard multiple imputation with proposed methods.
Standard imputation techniques are unbiased when data are missing at random (MAR).
Bias observed in standard techniques under missing not at random (MNAR) conditions.
Proposed TMLE methods remain unbiased even with model misspecification.

No takes yet. Share an insight, caveat, or question.

Observational methods estimate causal effects in mortality studies with missing exposure and confounders, suggesting better approaches.

Wen et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark