Foundations of Structural Causal Models with Cycles and Latent Variables

Abstract

Structural causal models (SCMs), also known as (nonparametric) structural equation models (SEMs), are widely used for causal modeling purposes. In particular, acyclic SCMs, also known as recursive SEMs, form a well-studied subclass of SCMs that generalize causal Bayesian networks to allow for latent confounders. In this paper, we investigate SCMs in a more general setting, allowing for the presence of both latent confounders and cycles. We show that in the presence of cycles, many of the convenient properties of acyclic SCMs do not hold in general: they do not always have a solution; they do not always induce unique observational, interventional and counterfactual distributions; a marginalization does not always exist, and if it exists the marginal model does not always respect the latent projection; they do not always satisfy a Markov property; and their graphs are not always consistent with their causal semantics. We prove that for SCMs in general each of these properties does hold under certain solvability conditions. Our work generalizes results for SCMs with cycles that were only known for certain special cases so far. We introduce the class of simple SCMs that extends the class of acyclic SCMs to the cyclic setting, while preserving many of the convenient properties of acyclic SCMs. With this paper we aim to provide the foundations for a general theory of statistical causal modeling with SCMs.

75 pages (including supplementary material)

References in corpus (24)

• Estimating high-dimensional intervention effects from observational data
• Distinguishing cause from effect using observational data: methods and benchmarks
• CAM: Causal additive models, high-dimensional order search and penalized regression
• Strong Completeness and Faithfulness in Bayesian Networks
• Directed Cyclic Graphical Representations of Feedback Models
• A Discovery Algorithm for Directed Cyclic Graphs
• Joint Causal Inference from Multiple Contexts
• Graphs for margins of Bayesian networks
• Causal Consistency of Structural Equation Models
• Discovering Cyclic Causal Models by Independent Components Analysis
• Margins of discrete Bayesian networks
• A Constraint Propagation Approach to Probabilistic Reasoning
• Learning stable and predictive structures in kinetic systems: Benefits of a causal approach
• From Ordinary Differential Equations to Structural Causal Models: the deterministic case
• On Deducing Conditional Independence from d-Separation in Causal Graphs with Feedback (Research Note)
• Discovering Cyclic Causal Models with Latent Variables: A General SAT-Based Procedure
• Markov Properties for Graphical Models with Cycles and Latent Variables
• Half-trek criterion for generic identifiability of linear structural equation models
• Markov equivalence of marginalized local independence graphs
• Cyclic Causal Discovery from Continuous Equilibrium Data
• Constraint-based Causal Discovery for Non-Linear Structural Causal Models with Cycles and Latent Confounders
• Beyond Structural Causal Models: Causal Constraints Models
• Causal Calculus in the Presence of Cycles, Latent Confounders and Selection Bias
• Constraint-Based Causal Discovery using Partial Ancestral Graphs in the presence of Cycles

Cited by in corpus (11)

• Latent Variable Models in the Era of Industrial Big Data: Extension and Beyond
• Causal Structure Learning: a Combinatorial Perspective
• What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective
• A general framework for cyclic and fine-tuned causal models and their compatibility with space-time
• Embedding cyclic information-theoretic structures in acyclic spacetimes: no-go results for indefinite causality
• Reasoning about Causality in Games
• Impossibility of superluminal signalling in Minkowski space-time does not rule out causal loops
• Model-based causal feature selection for general response types
• Relating Wigner's Friend Scenarios to Nonclassical Causal Compatibility, Monogamy Relations, and Fine Tuning
• Unfair Utilities and First Steps Towards Improving Them
• Spatio-Temporal Graphical Counterfactuals: An Overview