B0555
Title: Partial identification with noisy covariates: A robust optimization approach
Authors: Yixin Wang - University of Michigan (United States) [presenting]
Abstract: Causal inference from observational datasets often relies on measuring and adjusting for covariates. In practice, measurements of the covariates can often be noisy and/or biased, or only measurements of their proxies may be available. Directly adjusting for these imperfect measurements of the covariates can lead to biased causal estimates. Moreover, without additional assumptions, the causal effects are not point-identifiable due to the noise in these measurements. The partial identification of causal effects given noisy covariates are studied, under a user-specified assumption on the noise level. The key observation is that the identification of the average treatment effects (ATE) can be formulated as a robust optimization problem. This formulation leads to an efficient robust optimization algorithm that bounds the ATE with noisy covariates. It is shown that this robust optimization approach can extend a wide range of causal adjustment methods to perform partial identification, including backdoor adjustment, inverse propensity score weighting, double machine learning, and front door adjustment. Across synthetic and real datasets, it is found that this approach provides ATE bounds with a higher coverage probability than existing methods.
