Estimating average treatment effects when treatment data are absent in a target study
By: Lan Wen, Aaron L Sarvet
Potential Business Impact:
Helps understand what works when data is missing.
Researchers are frequently interested in understanding the causal effect of treatment interventions. However, in some cases, the treatment of interest--readily available in a randomized controlled trial (RCT)--is either not directly measured or entirely unavailable in observational datasets. This challenge has motivated the development of stochastic incremental propensity score interventions which operate on post-treatment exposures affected by the treatment of interest with the aim of approximating the causal effects of the treatment intervention. Yet, a key challenge lies in the fact that the precise distributional shift of these post-treatment exposures induced by the treatment is typically unknown, making it uncertain whether the approximation truly reflects the causal effect of interest. The primary objective of this paper is to explore data integration methodologies to characterize a distribution of post-treatment exposures resulting from the treatment in an external dataset, and to use this information to estimate counterfactual mean outcomes under treatment interventions, in settings where the observational data lack treatment information and the external data may not contain measurements of the outcome of interest. We will discuss the underlying assumptions required for this approach and provide methodological guidance on estimation strategies to address these challenges.
Similar Papers
Interpolated stochastic interventions based on propensity scores, target policies and treatment-specific costs
Methodology
Helps plan experiments with limited money.
Characterization and Learning of Causal Graphs with Latent Confounders and Post-treatment Selection from Interventional Data
Machine Learning (CS)
Finds true causes even with hidden factors.
Difference-in-differences with stochastic policy shifts of continuous treatments
Methodology
Helps measure how much jobs change from new rules.