Bounding the Effects of Continuous Treatments for Hidden Confounders

  title={Bounding the Effects of Continuous Treatments for Hidden Confounders},
  author={Myrl G. Marmarelis and Greg Ver Steeg and A. G. Galstyan},
Observational studies often seek to infer the causal effect of a treatment even though both the assigned treatment and the outcome depend on other confounding variables. An effective strategy for deal-ing with confounders is to estimate a propensity model that corrects for the relationship between covariates and assigned treatment. Unfortunately, the confounding variables themselves are not always observed, in which case we can only bound the propensity, and therefore bound the magnitude of… 

Figures and Tables from this paper


Bounds on the conditional and average treatment effect with unobserved confounding factors
A loss minimization approach that quantifies bounds on the conditional average treatment effect (CATE) when unobserved confounder have a bounded effect on the odds of treatment selection and a semi-parametric framework that extends/bounds the augmented inverse propensity weighted (AIPW) estimator for the ATE beyond the assumption that all confounders are observed.
Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
A new parametric interval estimator suited for highdimensional data, that estimates a range of possible CATE values when given a predefined bound on the level of hidden confounding, and incorporates model uncertainty so that practitioners can be made aware of such out-of-distribution data.
Interval Estimation of Individual-Level Causal Effects Under Unobserved Confounding
A functional interval estimator is developed that predicts bounds on the individual causal effects under realistic violations of unconfoundedness and is proved that it converges exactly to the tightest bounds possible on CATE when there may be unobserved confounders.
Scalable Sensitivity and Uncertainty Analysis for Causal-Effect Estimates of Continuous-Valued Interventions
A continuous treatment-effect marginal sensitivity model (CMSM) is developed and bounds that agree with both the observed data and a researcher-defined level of hidden confounding are derived, using a scalable algorithm to derive the bounds and uncertainty-aware deep models to estimate these bounds for high-dimensional, large-sample observational data.
Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome
This paper proposes a simple technique for assessing the range of plausible causal con- clusions from observational studies with a binary outcome and an observed categorical covariate. The technique
An Introduction to Proximal Causal Learning
A formal potential outcome framework for proximal causal learning is introduced, which while explicitly acknowledging covariate measurements as imperfect proxies of confounding mechanisms, offers an opportunity to learn about causal effects in settings where exchangeability on the basis of measured covariates fails.
Causal Effect Inference with Deep Latent-Variable Models
This work builds on recent advances in latent variable modeling to simultaneously estimate the unknown latent space summarizing the confounders and the causal effect and shows its method is significantly more robust than existing methods, and matches the state-of-the-art on previous benchmarks focused on individual treatment effects.
Conformal Sensitivity Analysis for Individual Treatment Effects
Estimating an individual treatment effect (ITE) is essential to personalized decision making. However, existing methods for estimating the ITE often rely on unconfoundedness, an assumption that is
Multiply Robust Causal Mediation Analysis with Continuous Treatments
This work extends the influence function-based estimator of Tchetgen Tchet Gen and Shpitser (2012) to deal with continuous treatments by utilizing a kernel smoothing approach and preserves the multiple robustness property of the estimator.
Sense and Sensitivity Analysis: Simple Post-Hoc Analysis of Bias Due to Unobserved Confounding
Austen plots are developed, a sensitivity analysis tool to aid domain experts' judgments about whether strong confounders in an observational study are plausible, by making it easier to reason about potential bias induced by unobserved confounding.