Local Discovery by Partitioning: Polynomial-Time Causal Discovery Around Exposure-Outcome Pairs

Jacqueline R. M. A. Maasch, Weishen Pan, Shantanu Gupta, Volodymyr Kuleshov, Kyra Gan, Fei Wang

Keywords: causal discovery, graphical models, covariate adjustment

TL;DR: Local causal discovery enables efficient valid adjustment set identification under limited prior knowledge and causal insufficiency.

Abstract: Causal discovery is crucial for causal inference in observational studies, as it can enable the identification of *valid adjustment sets* (VAS) for unbiased effect estimation. However, global causal discovery is notoriously hard in the nonparametric setting, with exponential time and sample complexity in the worst case. To address this, we propose *local discovery by partitioning* (LDP): a local causal discovery method that is tailored for downstream inference tasks without requiring parametric and pretreatment assumptions. LDP is a constraint-based procedure that returns a VAS for an exposure-outcome pair under latent confounding, given sufficient conditions. The total number of independence tests performed is worst-case quadratic with respect to the cardinality of the variable set. Asymptotic theoretical guarantees are numerically validated on synthetic graphs. Adjustment sets from LDP yield less biased and more precise average treatment effect estimates than baseline discovery algorithms, with LDP outperforming on confounder recall, runtime, and test count for VAS discovery. Notably, LDP ran at least $1300\times$ faster than baselines on a benchmark.

List Of Authors: Maasch, Jacqueline and Pan, Weishen and Gupta, Shantanu and Kuleshov, Volodymyr and Gan, Kyra and Wang, Fei

Code Url: https://github.com/jmaasch/ldp

Submission Number: 561