On model selection and model misspecification in causal inference

DOI10.1177/0962280210387717zbMath1365.62431OpenAlexW2107263285WikidataQ49083477 ScholiaQ49083477MaRDI QIDQ5272528

Stijn Vansteelandt, Maarten Bekaert, Gerda Claeskens

Publication date: 30 June 2017

Published in: Statistical Methods in Medical Research (Search for Journal in Brave)

Full work available at URL: https://lirias.kuleuven.be/handle/123456789/277522

zbMATH Keywords

model selectionmodel uncertaintycausal inferencedouble robustnesspropensity scoreconfounder-selectioninfluential weights

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10)

Related Items

Robust inference using hierarchical likelihood approach for heavy-tailed longitudinal outcomes with missing data: an alternative to inverse probability weighted generalized estimating equations, Model averaged double robust estimation, Model averaging in semiparametric estimation of treatment effects, Data-driven algorithms for dimension reduction in causal inference, Variable selection – A review and recommendations for the practicing statistician, Assessing mediating effects of high‐dimensional microbiome measurements in dietary intervention studies, A semiparametric method for evaluating causal effects in the presence of error‐prone covariates, Power and sample size for observational studies of point exposure effects, Bayesian data fusion: Probabilistic sensitivity analysis for unmeasured confounding using informative priors based on secondary data, Making apples from oranges: Comparing noncollapsible effect estimators and their standard errors after adjustment for different covariate sets, A causal exposure response function with local adjustment for confounding: estimating health effects of exposure to low levels of ambient fine particulate matter, Bandwidth selection for statistical matching and prediction, Neighborhood-based cross fitting approach to treatment effects with high-dimensional data, The role of exchangeability in causal inference, A method for increasing the robustness of multiple imputation, A High‐dimensional Focused Information Criterion, Discussions, Stratified doubly robust estimators for the average causal effect, Efficient estimation of the distribution of time to composite endpoint when some endpoints are only partially observed, Correlation and efficiency of propensity score-based estimators for average causal effects, High-dimensional confounding adjustment using continuous Spike and Slab priors, Estimation of Direct Effects for Survival Data by using the Aalen Additive Hazards Model, Covariate selection with group lasso and doubly robust estimation of causal effects, Doubly robust policy evaluation and optimization, Structural nested models and G-estimation: the partially realized promise

Cites Work

• Unnamed Item
• Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data
• Comment: Performance of double-robust estimators when ``inverse probability weights are highly variable
• Comment: Understanding OR, PS and DR
• A semiparametric model selection criterion with applications to the marginal structural model
• Confounding and collapsibility in causal inference
• A cross-validation deletion-substitution-addition model selection algorithm: application to marginal structural models
• Semiparametric theory and missing data.
• Adjustment uncertainty in effect estimation
• Using Regression Models to Analyze Randomized Trials: Asymptotically Valid Hypothesis Tests Despite Incorrectly Specified Models
• Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data
• Semiparametric Estimation of Treatment Effect in a Pretest‐Posttest Study
• Model Selection and Model Averaging
• The central role of the propensity score in observational studies for causal effects
• A Note About Models for Selectivity Bias
• Frequentist Model Average Estimators
• The Focused Information Criterion
• Estimating Exposure Effects by Modelling the Expectation of Exposure Conditional on Confounders
• A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect
• The Method of Path Coefficients
• Variable Selection for Logistic Regression Using a Prediction‐Focused Information Criterion
• Regularization and Variable Selection Via the Elastic Net
• Multiply Robust Inference for Statistical Interactions