Nested Markov Properties for Acyclic Directed Mixed Graphs

Abstract

Conditional independence models associated with directed acyclic graphs (DAGs) may be characterized in at least three different ways: via a factorization, the global Markov property (given by the d-separation criterion), and the local Markov property. Marginals of DAG models also imply equality constraints that are not conditional independences; the well-known ``Verma constraint'' is an example. Constraints of this type are used for testing edges, and in a computationally efficient marginalization scheme via variable elimination. We show that equality constraints like the ``Verma constraint'' can be viewed as conditional independences in kernel objects obtained from joint distributions via a fixing operation that generalizes conditioning and marginalization. We use these constraints to define, via ordered local and global Markov properties, and a factorization, a graphical model associated with acyclic directed mixed graphs (ADMGs). We prove that marginal distributions of DAG models lie in this model, and that a set of these constraints given by Tian provides an alternative definition of the model. Finally, we show that the fixing operation used to define the model leads to a particularly simple characterization of identifiable causal effects in hidden variable causal DAG models.

36 pages (not including appendix and references), 9 figures. Fixed a definition following equation (16) in the main text (the fix is shown in blue text). Fixed double parentheses showing up for some references

References in corpus (10)

• The Inflation Technique for Causal Inference with Latent Variables
• On the Testability of Causal Models with Latent and Instrumental Variables
• Markov equivalence for ancestral graphs
• Estimation of Effects of Sequential Treatments by Reparameterizing Directed Acyclic Graphs
• Probability distributions with summary graph structure
• A Complete Generalized Adjustment Criterion
• On the Testable Implications of Causal Models with Hidden Variables
• Differentiable Causal Discovery Under Unmeasured Confounding
• Maximum likelihood fitting of acyclic directed mixed graphs to binary data
• An Efficient Algorithm for Computing Interventional Distributions in Latent Variable Causal Models

Cited by in corpus (31)

• Constraint-based Causal Discovery from Multiple Interventions over Overlapping Variable Sets
• Causal Compatibility Inequalities Admitting Quantum Violations in the Triangle Structure
• The Inflation Technique Completely Solves the Causal Compatibility Problem
• Markov equivalence of marginalized local independence graphs
• On efficient adjustment in causal graphs
• Evaluating Model Robustness and Stability to Dataset Shift
• Differentiable Causal Discovery Under Unmeasured Confounding
• Parameter and Structure Learning in Nested Markov Models
• Quantum Causal Models
• Preventing Failures Due to Dataset Shift: Learning Predictive Models That Transport
• 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
• Model selection and local geometry
• A review of Gaussian Markov models for conditional independence
• Rank-transformed subsampling: inference for multiple data splitting and exchangeable p-values
• Conditional independence ideals with hidden variables
• Efficient adjustment sets in causal graphical models with hidden variables
• Switching Regression Models and Causal Inference in the Presence of Discrete Latent Variables
• A Combinatorial Solution to Causal Compatibility
• Identifying causal effects in maximally oriented partially directed acyclic graphs
• Entropic Inequality Constraints from $e$-separation Relations in Directed Acyclic Graphs with Hidden Variables
• Latent-free equivalent mDAGs
• Learning latent causal graphs via mixture oracles
• Distributional Invariances and Interventional Markov Equivalence for Mixed Graph Models
• General Identification of Dynamic Treatment Regimes Under Interference
• The neighborhood lattice for encoding partial correlations in a Hilbert space
• Towards Characterising Bayesian Network Models under Selection
• Path Dependent Structural Equation Models
• The Proximal ID Algorithm
• A Kernel Test for Causal Association via Noise Contrastive Backdoor Adjustment