arXiv subjects

Edward H Kennedy

Publications and source records attributed to Edward H Kennedy.

3 recordsLinked to original sources

arXiv2020

Sensitivity analysis via the proportion of unmeasured confounding

In observational studies, identification of ATEs is generally achieved by assuming that the correct set of confounders has been measured and properly included in the relevant models. Because this assumption is both strong and untestable, a sensitivity analysis should be performed. Common approaches include modeling the bias directly or varying the propensity scores to probe the effects of a potential unmeasured confounder. In this paper, we take a novel approach whereby the sensitivity parameter is the "proportion of unmeasured confounding:" the proportion of units for whom the treatment is not as good as randomized even after conditioning on the observed covariates. We consider different assumptions on the probability of a unit being unconfounded. In each case, we derive sharp bounds on the average treatment effect as a function of the sensitivity parameter and propose nonparametric estimators that allow flexible covariate adjustment. We also introduce a one-number summary of a study's robustness to the number of confounded units. Finally, we explore finite-sample properties via simulation, and apply the methods to an observational database used to assess the effects of right heart catheterization.

stat.ME↗

arXiv2020

Challenges in Obtaining Valid Causal Effect Estimates with Machine Learning Algorithms

Unlike parametric regression, machine learning (ML) methods do not generally require precise knowledge of the true data generating mechanisms. As such, numerous authors have advocated for ML methods to estimate causal effects. Unfortunately, ML algorithms can perform worse than parametric regression. We demonstrate the performance of ML-based single- and double-robust estimators. We use 100 Monte Carlo samples with sample sizes of 200, 1200, and 5000 to investigate bias and confidence interval coverage under several scenarios. In a simple confounding scenario, confounders were related to the treatment and the outcome via parametric models. In a complex confounding scenario, the simple confounders were transformed to induce complicated nonlinear relationships. In the simple scenario, when ML algorithms were used, double-robust estimators were superior to single-robust estimators. In the complex scenario, single-robust estimators with ML algorithms were at least as biased as estimators using misspecified parametric models. Double-robust estimators were less biased, but coverage was well below nominal. The use of sample splitting, inclusion of confounder interactions, reliance on a richly specified ML algorithm, and use of doubly robust estimators was the only explored approach that yielded negligible bias and nominal coverage. Our results suggest that ML based singly robust methods should be avoided.

stat.ME↗

arXiv2019

Instrumental Variable Methods using Dynamic Interventions

Recent work on dynamic interventions has greatly expanded the range of causal questions researchers can study while weakening identifying assumptions and yielding effects that are more practically relevant. However, most work in dynamic interventions to date has focused on settings where we directly alter some unconfounded treatment of interest. In policy analysis, decision makers rarely have this level of control over behaviors or access to experimental data. Instead, they are often faced with treatments they can affect only indirectly and whose effects must be learned from observational data. In this paper, we propose new estimands and estimators of causal effects based on dynamic interventions with instrumental variables. This method does not rely on parametric models and does not require an experiment. Instead, we estimate the effect of a dynamic intervention on the instrument. This robustness should reassure policy makers that these estimates can be used to effectively inform policy. We demonstrate the usefulness of this estimation strategy in a case study examining the effect of visitation on recidivism.

stat.ME↗