Surrogate Outcomes and Transportability
This work addresses a fundamental challenge in causal inference for researchers, offering a theoretical extension to improve identifiability in mixed data settings, though it appears incremental as it generalizes existing concepts like surrogate experiments.
The paper tackles the problem of identifying causal effects when experimental and observational data are insufficient individually, by measuring surrogate outcomes in experiments, and shows that transportability provides a sufficient criterion for identifiability in a large class of queries.
Identification of causal effects is one of the most fundamental tasks of causal inference. We consider an identifiability problem where some experimental and observational data are available but neither data alone is sufficient for the identification of the causal effect of interest. Instead of the outcome of interest, surrogate outcomes are measured in the experiments. This problem is a generalization of identifiability using surrogate experiments and we label it as surrogate outcome identifiability. We show that the concept of transportability provides a sufficient criteria for determining surrogate outcome identifiability for a large class of queries.