Causal inference with confounders missing not at random
From MaRDI portal
Publication:5212911
DOI10.1093/BIOMET/ASZ048zbMATH Open1435.62059arXiv1702.03951OpenAlexW2976627433WikidataQ127354565 ScholiaQ127354565MaRDI QIDQ5212911FDOQ5212911
Authors:
Publication date: 30 January 2020
Published in: Biometrika (Search for Journal in Brave)
Abstract: It is important to draw causal inference from observational studies, which, however, becomes challenging if the confounders have missing values. Generally, causal effects are not identifiable if the confounders are missing not at random. We propose a novel framework to nonparametrically identify causal effects with confounders subject to an outcome-independent missingness, that is, the missing data mechanism is independent of the outcome, given the treatment and possibly missing confounders. We then propose a nonparametric two-stage least squares estimator and a parametric estimator for causal effects.
Full work available at URL: https://arxiv.org/abs/1702.03951
Recommendations
- Causal inference: a missing data perspective
- Treatment effects estimation with missing not at random data without outcome modeling
- scientific article; zbMATH DE number 7366118
- A robust and efficient approach to causal inference based on sparse sufficient dimension reduction
- Identification of causal effects in the presence of nonignorable missing outcome values
identifiabilityintegral equationcompletenesstwo-stage least squares estimatorill-posed inverse problemoutcome-independent missingness
Cited In (27)
- Causal inference: a missing data perspective
- Causal inference with missingness in confounder
- Doubly weighted mean score estimating functions with a partially observed effect modifier
- Causal Inference Without Counterfactuals
- To Adjust or not to Adjust? Estimating the Average Treatment Effect in Randomized Experiments with Missing Covariates
- A self-censoring model for multivariate nonignorable nonmonotone missing data
- Matrix completion under complex survey sampling
- Treatment effects estimation with missing not at random data without outcome modeling
- Combining multiple observational data sources to estimate causal effects
- A unified inference framework for multiple imputation using martingales
- Measurement bias and effect restoration in causal inference
- The role of covariate balance in observational studies
- Causal inference with observational data under cluster-specific non-ignorable assignment mechanism
- Doubly robust treatment effect estimation with missing attributes
- Non-causality: The role of the omitted variables
- Nonparametric tests for treatment effect heterogeneity in observational studies
- A Review of Spatial Causal Inference Methods for Environmental and Epidemiological Applications
- Causal inference accounting for unobserved confounding after outcome regression and doubly robust estimation
- Nested doubly robust estimating equations for causal analysis with an incomplete effect modifier
- Assessing the causal effect of binary interventions from observational panel data with few treated units
- Estimating treatment effects with partially observed covariates using outcome regression with missing indicators
- Multiply robust estimation of the average treatment effect with missing outcomes
- Nonparametric causal inference with confounders missing not at random
- Generalizing treatment effects with incomplete covariates: identifying assumptions and multiple imputation algorithms
- Assessing lack of common support in causal inference using Bayesian nonparametrics: Implications for evaluating the effect of breastfeeding on children's cognitive outcomes
- Semiparametric Inference of Causal Effect with Nonignorable Missing Confounders
- Combining Parametric and Nonparametric Models to Estimate Treatment Effects in Observational Studies
This page was built for publication: Causal inference with confounders missing not at random
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5212911)
