Local Causal Discovery for Estimating Causal Effects
Shantanu Gupta, David Childers, Zachary C. Lipton
TL;DR
The paper addresses identifying the set of possible ATE values, $\Theta^{*}$, when the causal graph is unknown by exploiting local structure around the treatment. It introduces Local Discovery using Eager Collider Checks (LDECC), a local causal discovery method that orients the treatment's parents via ECCs and Minimal Neighbor Separators, with correctness guaranteed under the CFA. It contrasts LDECC with existing local methods (e.g., SD) in terms of computational and faithfulness requirements, arguing for complementary use and potential hybrids to broaden efficient applicability. Empirical results on synthetic and semi-synthetic graphs show that LDECC achieves competitive accuracy and recall while using fewer CI tests, supporting its practical value for efficient local causal discovery and conservative ATE-set identification.
Abstract
Even when the causal graph underlying our data is unknown, we can use observational data to narrow down the possible values that an average treatment effect (ATE) can take by (1) identifying the graph up to a Markov equivalence class; and (2) estimating that ATE for each graph in the class. While the PC algorithm can identify this class under strong faithfulness assumptions, it can be computationally prohibitive. Fortunately, only the local graph structure around the treatment is required to identify the set of possible ATE values, a fact exploited by local discovery algorithms to improve computational efficiency. In this paper, we introduce Local Discovery using Eager Collider Checks (LDECC), a new local causal discovery algorithm that leverages unshielded colliders to orient the treatment's parents differently from existing methods. We show that there exist graphs where LDECC exponentially outperforms existing local discovery algorithms and vice versa. Moreover, we show that LDECC and existing algorithms rely on different faithfulness assumptions, leveraging this insight to weaken the assumptions for identifying the set of possible ATE values.