Addressing positivity violations in causal effect estimation using Gaussian process priors
- 6 November 2022
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 42 (1) , 33-51
Abstract
In observational studies, causal inference relies on several key identifying assumptions. One identifiability condition is the positivity assumption, which requires the probability of treatment be bounded away from 0 and 1. That is, for every covariate combination, it should be possible to observe both treated and control subjects the covariate distributions should overlap between treatment arms. If the positivity assumption is violated, population‐level causal inference necessarily involves some extrapolation. Ideally, a greater amount of uncertainty about the causal effect estimate should be reflected in such situations. With that goal in mind, we construct a Gaussian process model for estimating treatment effects in the presence of practical violations of positivity. Advantages of our method include minimal distributional assumptions, a cohesive model for estimating treatment effects, and more uncertainty associated with areas in the covariate space where there is less overlap. We assess the performance of our approach with respect to bias and efficiency using simulation studies. The method is then applied to a study of critically ill female patients to examine the effect of undergoing right heart catheterization.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Optimal Matching of an Optimally Chosen Subset in Observational StudiesJournal of Computational and Graphical Statistics, 2012
- Variable Selection for Nonparametric Gaussian Process Priors: Models and Computational StrategiesStatistical Science, 2011
- Diagnosing and responding to violations in the positivity assumptionStatistical Methods in Medical Research, 2010
- Invited Commentary: Positivity in PracticeAmerican Journal of Epidemiology, 2010
- The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trialsStatistics in Medicine, 2006
- Estimating causal effects from epidemiological dataJournal of Epidemiology and Community Health, 2006
- The Dangers of Extreme CounterfactualsPolitical Analysis, 2006
- Causal Inference Using Potential OutcomesJournal of the American Statistical Association, 2005
- Seeking efficient data augmentation schemes via conditional and marginal augmentationBiometrika, 1999
- Bayesian Analysis of Binary and Polychotomous Response DataJournal of the American Statistical Association, 1993