Pearl's Calculus of Intervention Is Complete

Abstract

This paper is concerned with graphical criteria that can be used to solve the problem of identifying casual effects from nonexperimental data in a causal Bayesian network structure, i.e., a directed acyclic graph that represents causal relationships. We first review Pearl's work on this topic [Pearl, 1995], in which several useful graphical criteria are presented. Then we present a complete algorithm [Huang and Valtorta, 2006b] for the identifiability problem. By exploiting the completeness of this algorithm, we prove that the three basic do-calculus rules that Pearl presents are complete, in the sense that, if a causal effect is identifiable, there exists a sequence of applications of the rules of the do-calculus that transforms the causal effect formula into a formula that only includes observational quantities.

Appears in Proceedings of the Twenty-Second Conference on Uncertainty in Artificial Intelligence (UAI2006)

References in corpus (3)

• Probabilistic Evaluation of Sequential Plans from Causal Models with Hidden Variables
• Testing Identifiability of Causal Effects
• On the Testable Implications of Causal Models with Hidden Variables

Cited by in corpus (33)

• A General Algorithm for Deciding Transportability of Experimental Results
• Separators and Adjustment Sets in Causal Graphs: Complete Criteria and an Algorithmic Framework
• Causal Inference in Travel Demand Modeling (and the lack thereof)
• The Do-Calculus Revisited
• Causal Imitation Learning with Unobserved Confounders
• Study design in causal models
• Causal Reinforcement Learning using Observational and Interventional Data
• Differentiable Causal Discovery Under Unmeasured Confounding
• Membership Inference Attacks and Generalization: A Causal Perspective
• Causal Effect Identification from Multiple Incomplete Data Sources: A General Search-based Approach
• Counterfactual analyses with graphical models based on local independence
• Causal Inference in the Presence of Interference in Sponsored Search Advertising
• Causal Calculus in the Presence of Cycles, Latent Confounders and Selection Bias
• Semiparametric Inference For Causal Effects In Graphical Models With Hidden Variables
• The Importance of Causality in Decision Making: A Perspective on Recommender Systems
• Towards Causality-Aware Inferring: A Sequential Discriminative Approach for Medical Diagnosis
• On Identifying Total Effects in the Presence of Latent Variables and Selection bias
• Surrogate Outcomes and Transportability
• Maximum likelihood fitting of acyclic directed mixed graphs to binary data
• Estimation of causal effects with small data in the presence of trapdoor variables
• Probabilistic Reasoning across the Causal Hierarchy
• Identifying causal effects in maximally oriented partially directed acyclic graphs
• Replacing the do-calculus with Bayes rule
• Causal Transportability of Experiments on Controllable Subsets of Variables: z-Transportability
• Identifying Counterfactual Queries with the R Package cfid
• Causal Effect Identification with Context-specific Independence Relations of Control Variables
• Causal Inference for Influence Propagation -- Identifiability of the Independent Cascade Model
• Reconciling Causality and Statistics
• Causal Inference in medicine and in health policy, a summary
• Causality: a decision theoretic approach
• Causal Effect Identification in Acyclic Directed Mixed Graphs and Gated Models
• Learning Instrumental Variables with Non-Gaussianity Assumptions: Theoretical Limitations and Practical Algorithms
• Estimating complex causal effects from incomplete observational data