Unveiling the Structure of Do-Calculus Reasoning via Derivation Graphs
For researchers in causal inference, this work provides a systematic framework to navigate the do-calculus reasoning space, potentially improving estimator efficiency.
The paper introduces derivation graphs to characterize the full space of equivalent interventional expressions under do-calculus, showing that at most four rule applications are needed and that multiple valid estimands can yield more efficient estimators.
The do-calculus defines a general system of inference for interventional queries, allowing causal quantities to be transformed through successive applications of its rules. This process induces a rich space of equivalent interventional expressions, but combining and ordering these rules remains challenging. In this work, we introduce derivation graphs, which represent how do-calculus rules are applied and combined, and characterize the full space of observational and interventional probabilities which are equivalent under the do-calculus. The structure of these graphs yields a simple procedure that uses at most four applications of do-calculus rules. Finally, we show how applying identification algorithms to equivalent causal queries produces multiple valid estimands for the same causal quantity, eventually yielding more efficient estimators.