Identifiability of Gaussian structural equation models with equal error variances

Abstract

We consider structural equation models in which variables can be written as a function of their parents and noise terms, which are assumed to be jointly independent. Corresponding to each structural equation model, there is a directed acyclic graph describing the relationships between the variables. In Gaussian structural equation models with linear functions, the graph can be identified from the joint distribution only up to Markov equivalence classes, assuming faithfulness. In this work, we prove full identifiability if all noise variables have the same variances: the directed acyclic graph can be recovered from the joint Gaussian distribution. Our result has direct implications for causal inference: if the data follow a Gaussian structural equation model with equal error variances and assuming that all variables are observed, the causal structure can be inferred from observational data only. We propose a statistical method and an algorithm that exploit our theoretical findings.

References in corpus (4)

• A Transformational Characterization of Equivalent Bayesian Network Structures
• Geometry of the faithfulness assumption in causal inference
• ${\ell }_0$-penalized maximum likelihood for sparse directed acyclic graphs
• Identifiability of Causal Graphs using Functional Models

Cited by in corpus (61)

• Distinguishing cause from effect using observational data: methods and benchmarks
• CAM: Causal additive models, high-dimensional order search and penalized regression
• Causal Discovery with Continuous Additive Noise Models
• ${\ell }_0$-penalized maximum likelihood for sparse directed acyclic graphs
• Causal Discovery with Reinforcement Learning
• High-dimensional learning of linear causal networks via inverse covariance estimation
• Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
• On the Role of Sparsity and DAG Constraints for Learning Linear DAGs
• On Causal Discovery with Equal Variance Assumption
• Concave Penalized Estimation of Sparse Gaussian Bayesian Networks
• Causality in cognitive neuroscience: concepts, challenges, and distributional robustness
• Causal Structure Learning: a Combinatorial Perspective
• Causal interpretation of stochastic differential equations
• Causal Fourier Analysis on Directed Acyclic Graphs and Posets
• Score-based Causal Learning in Additive Noise Models
• Learning Sparse Nonparametric DAGs
• Distributed Learning of Generalized Linear Causal Networks
• Estimation of Structural Causal Model via Sparsely Mixing Independent Component Analysis
• Nonlinear Causal Discovery with Confounders
• Gradient-Based Neural DAG Learning
• Learning Causal Structures Using Regression Invariance
• A Local Method for Identifying Causal Relations under Markov Equivalence
• Beware of the Simulated DAG! Causal Discovery Benchmarks May Be Easy To Game
• Discussion of Big Bayes Stories and BayesBag
• Towards Federated Bayesian Network Structure Learning with Continuous Optimization
• Quantifying identifiability in independent component analysis
• Confidence in Causal Inference under Structure Uncertainty in Linear Causal Models with Equal Variances
• Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values
• Ordering-Based Causal Discovery with Reinforcement Learning
• Model-based causal feature selection for general response types
• Learning Linear Non-Gaussian Causal Models in the Presence of Latent Variables
• Causal Inference from Slowly Varying Nonstationary Processes
• On the Convergence of Continuous Constrained Optimization for Structure Learning
• Efficient least squares for estimating total effects under linearity and causal sufficiency
• Active Invariant Causal Prediction: Experiment Selection through Stability
• Complexity analysis of Bayesian learning of high-dimensional DAG models and their equivalence classes
• DYNOTEARS: Structure Learning from Time-Series Data
• A polynomial-time algorithm for learning nonparametric causal graphs
• Discovery and inference of a causal network with hidden confounding
• Marginal integration for nonparametric causal inference
• Structure learning in polynomial time: Greedy algorithms, Bregman information, and exponential families
• Learning latent causal graphs via mixture oracles
• Confidence in Causal Discovery with Linear Causal Models
• Optimization-based Causal Estimation from Heterogenous Environments
• Optimal Experiment Design for Causal Discovery from Fixed Number of Experiments
• Efficient Bayesian network structure learning via local Markov boundary search
• Identifiability of Gaussian Structural Equation Models with Homogeneous and Heterogeneous Error Variances
• Causal Link Discovery with Unequal Edge Error Tolerance
• Typing assumptions improve identification in causal discovery
• Independent mechanism analysis, a new concept?
• The Effect of Noise Level on Causal Identification with Additive Noise Models
• Learning Joint Nonlinear Effects from Single-variable Interventions in the Presence of Hidden Confounders
• Causality and Generalizability: Identifiability and Learning Methods
• On the Role of Entropy-based Loss for Learning Causal Structures with Continuous Optimization
• Gradient-based Causal Structure Learning with Normalizing Flow
• Testing Mediation Effects Using Logic of Boolean Matrices
• Causal Discovery in Linear Structural Causal Models with Deterministic Relations
• Colored Gaussian directed acyclic graphical models
• Identifiability of AMP chain graph models
• polyDAG: Polynomial Acyclicity Constraints for Efficient Continuous Causal Discovery in Visual Semantic Graphs
• Causal network learning with non-invertible functional relationships