Sensitivity of weighted least squares estimators to omitted variables

TL;DR

This paper develops sensitivity analysis tools for weighted least squares estimators to unobserved confounding, providing bounds and diagnostics to assess the robustness of causal effect estimates in observational studies.

Contribution

It introduces a framework using weighted partial R^2 parameters to quantify and bound the impact of unobserved confounding on weighted linear regression estimators.

Findings

Bias depends on two weighted partial R^2 parameters.

Derived formal bounds on unobserved confounding strength.

Tools applicable with any weighting scheme and no distributional assumptions.

Abstract

This paper introduces tools for assessing the sensitivity, to unobserved confounding, of a common estimator of the causal effect of a treatment on an outcome that employs weights: the weighted linear regression of the outcome on the treatment and observed covariates. We demonstrate through the omitted variable bias framework that the bias of this estimator is a function of two intuitive sensitivity parameters: (i) the proportion of weighted variance in the treatment that unobserved confounding explains given the covariates and (ii) the proportion of weighted variance in the outcome that unobserved confounding explains given the covariates and the treatment, i.e., two weighted partial $R^2$ values. Following previous work, we define sensitivity statistics that lend themselves well to routine reporting, and derive formal bounds on the strength of the unobserved confounding with (a multiple of) the strength of select dimensions of the covariates, which help the user determine if unobserved confounding that would alter one's conclusions is plausible. We also propose tools for adjusted inference. A key choice we make is to examine only how the (weighted) outcome model is influenced by unobserved confounding, rather than examining how the weights have been biased by omitted confounding. One benefit of this choice is that the resulting tool applies with any weights (e.g., inverse-propensity score, matching, or covariate balancing weights). Another benefit is that we can rely on simple omitted variable bias approaches that, for example, impose no distributional assumptions on the data or unobserved confounding, and can address bias from misspecification in the observed data. We make these tools available in the weightsense package for the R computing language.