Conditional Local Independence Testing with Application to Dynamic Causal Discovery
Mingzhou Liu, Xinwei Sun, Yizhou Wang
2025-06-09Causal Discovery
Abstract
In this note, we extend the conditional local independence testing theory developed in Christgau et al. (2024) to Ito processes. The result can be applied to causal discovery in dynamic systems.
Related Papers
Lower Bounds on the Size of Markov Equivalence Classes2025-06-26Causal discovery in deterministic discrete LTI-DAE systems2025-06-25Exploring Big Five Personality and AI Capability Effects in LLM-Simulated Negotiation Dialogues2025-06-19Time-Varying Home Field Advantage in Football: Learning from a Non-Stationary Causal Process2025-06-13Paths to Causality: Finding Informative Subgraphs Within Knowledge Graphs for Knowledge-Based Causal Discovery2025-06-10Permutation-Free High-Order Interaction Tests2025-06-06Nonlinear Causal Discovery for Grouped Data2025-06-05Causal Explainability of Machine Learning in Heart Failure Prediction from Electronic Health Records2025-06-03