High-dimensional confounding adjustment using continuous Spike and Slab priors

DOI10.1214/18-BA1131MaRDI QIDQ2316985zbMATH OpenOpenAlexWikidataFDO

Authors Joseph Antonelli, Giovanni Parmigiani, Francesca Dominici

Publication date 7 August 2019

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1704.07532, https://projecteuclid.org/euclid.ba/1560240029

zbMATH Keywords

causal inferencehigh-dimensional datashrinkage priorsBayesian variable selection

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10) Estimation in multivariate analysis (62H12)

Abstract: In observational studies, estimation of a causal effect of a treatment on an outcome relies on proper adjustment for confounding. If the number of the potential confounders ($p$) is larger than the number of observations ($n$), then direct control for all potential confounders is infeasible. Existing approaches for dimension reduction and penalization are generally aimed at predicting the outcome, and are less suited for estimation of causal effects. Under standard penalization approaches (e.g. Lasso), if a variable $X_j$ is strongly associated with the treatment $T$ but weakly with the outcome $Y$, the coefficient $$ will be shrunk towards zero thus leading to confounding bias. Under the assumption of a linear model for the outcome and sparsity, we propose continuous spike and slab priors on the regression coefficients $$ corresponding to the potential confounders $X_j$. Specifically, we introduce a prior distribution that does not heavily shrink to zero the coefficients ($$s) of the $X_j$s that are strongly associated with $T$ but weakly associated with $Y$. We compare our proposed approach to several state of the art methods proposed in the literature. Our proposed approach has the following features: 1) it reduces confounding bias in high dimensional settings; 2) it shrinks towards zero coefficients of instrumental variables; and 3) it achieves good coverages even in small sample sizes. We apply our approach to the National Health and Nutrition Examination Survey (NHANES) data to estimate the causal effects of persistent pesticide exposure on triglyceride levels.

Recommendations

• Negotiating multicollinearity with spike-and-slab priors
• A majorization-minimization approach to variable selection using spike and slab priors
• Spike and slab variable selection: frequentist and Bayesian strategies
• Dynamic variable selection with spike-and-slab process priors
• Spike-and-slab priors for function selection in structured additive regression models
• Fast Bayesian variable selection for high dimensional linear models: marginal solo spike and slab priors
• Theoretical and Computational Aspects of Continuous Spike-and-Slab Priors
• Bayesian estimation of sparse signals with a continuous spike-and-slab prior
• Bayesian inference for spatio-temporal spike-and-slab priors
• Recent Theoretical Advances with the Discrete Spike-and-Slab Priors

Cites work

• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• A significance test for the lasso
• Accounting for uncertainty in confounder and effect modifier selection when estimating average causal effects in generalized linear models
• Adjustment uncertainty in effect estimation
• Approximate residual balancing: debiased inference of average treatment effects in high dimensions
• Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
• Bayes and empirical-Bayes multiplicity adjustment in the variable-selection problem
• Bayesian Analysis of Binary and Polychotomous Response Data
• Bayesian data analysis.
• Bayesian effect estimation accounting for adjustment uncertainty
• Bayesian factorizations of big sparse tensors
• Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion)
• Confounder selection via penalized credible regions
• Covariate selection for the nonparametric estimation of an average treatment effect
• Dirichlet-Laplace priors for optimal shrinkage
• Empirical Bayes Gibbs sampling
• For objective causal inference, design trumps analysis
• High-dimensional confounding adjustment using continuous Spike and Slab priors
• Inference on treatment effects after selection among high-dimensional controls
• Model averaged double robust estimation
• On model selection and model misspecification in causal inference
• Outcome-adaptive Lasso: variable selection for causal inference
• Post-selection inference for \(\ell_1\)-penalized likelihood models
• Program evaluation and causal inference with high-dimensional data
• Regularization and Variable Selection Via the Elastic Net
• Robust inference on average treatment effects with possibly more covariates than observations
• Stable weights that balance covariates for estimation with incomplete outcome data
• The Adaptive Lasso and Its Oracle Properties
• The Bayesian Lasso
• The central role of the propensity score in observational studies for causal effects
• The horseshoe estimator for sparse signals
• The spike-and-slab LASSO
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties

Cited in

(13)

• Improved inference for doubly robust estimators of heterogeneous treatment effects
• Discussion on “Spatial+: a novel approach to spatial confounding” by Emiko Dupont, Simon N. Wood, and Nicole H. Augustin
• The how and why of Bayesian nonparametric causal inference
• A causal exposure response function with local adjustment for confounding: estimating health effects of exposure to low levels of ambient fine particulate matter
• Spike-and-Slab Group Lassos for Grouped Regression and Sparse Generalized Additive Models
• Mitigating unobserved spatial confounding when estimating the effect of supermarket access on cardiovascular disease deaths
• Causal inference in high dimensions: A marriage between Bayesian modeling and good frequentist properties
• High-dimensional confounding adjustment using continuous Spike and Slab priors
• Ultra-High Dimensional Variable Selection for Doubly Robust Causal Inference
• Comment: Regularization via Bayesian Penalty Mixing
• A Bayesian model of microbiome data for simultaneous identification of covariate associations and prediction of phenotypic outcomes
• Bayesian Approaches to Shrinkage and Sparse Estimation
• A new covariate selection strategy for high dimensional data in causal effect estimation with multivariate treatments

Describes a project that uses

Uses Software

• BayesDA
• covTest
• sbw

This page was built for publication: High-dimensional confounding adjustment using continuous Spike and Slab priors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2316985)