Causal Inference on Discrete Data using Additive Noise Models

Abstract

Inferring the causal structure of a set of random variables from a finite sample of the joint distribution is an important problem in science. Recently, methods using additive noise models have been suggested to approach the case of continuous variables. In many situations, however, the variables of interest are discrete or even have only finitely many states. In this work we extend the notion of additive noise models to these cases. We prove that whenever the joint distribution $\prob^{(X,Y)}$ admits such a model in one direction, e.g. $Y=f(X)+N, N \independent X$, it does not admit the reversed model $X=g(Y)+\tilde N, \tilde N \independent Y$ as long as the model is chosen in a generic way. Based on these deliberations we propose an efficient new algorithm that is able to distinguish between cause and effect for a finite sample of discrete variables. In an extensive experimental study we show that this algorithm works both on synthetic and real data sets.

References in corpus (2)

• On the Identifiability of the Post-Nonlinear Causal Model
• Identifying confounders using additive noise models

Cited by in corpus (30)

• Distinguishing cause from effect using observational data: methods and benchmarks
• Quantifying causal influences
• Causal Discovery with Continuous Additive Noise Models
• Causal inference via algebraic geometry: feasibility tests for functional causal structures with two binary observed variables
• A Sound and Complete Algorithm for Learning Causal Models from Relational Data
• Analysis of cause-effect inference by comparing regression errors
• Causal Inference on Discrete Data via Estimating Distance Correlations
• Consistency of Causal Inference under the Additive Noise Model
• Gradient-Based Neural DAG Learning
• An Information Criterion for Inferring Coupling in Distributed Dynamical Systems
• Learning directed acyclic graphs based on sparsest permutations
• Causal Discovery for Manufacturing Domains
• SADA: A General Framework to Support Robust Causation Discovery with Theoretical Guarantee
• Shared Causal Paths underlying Alzheimer's dementia and Type 2 Diabetes
• Estimating Causal Direction and Confounding of Two Discrete Variables
• Robust Learning via Cause-Effect Models
• Combining Linear Non-Gaussian Acyclic Model with Logistic Regression Model for Estimating Causal Structure from Mixed Continuous and Discrete Data
• On Azadkia-Chatterjee's conditional dependence coefficient
• Merging joint distributions via causal model classes with low VC dimension
• Causal Discovery in a Binary Exclusive-or Skew Acyclic Model: BExSAM
• Bayesian estimation of possible causal direction in the presence of latent confounders using a linear non-Gaussian acyclic structural equation model with individual-specific effects
• Accelerating Recursive Partition-Based Causal Structure Learning
• Causal Order Identification to Address Confounding: Binary Variables
• A Subsampling-Based Method for Causal Discovery on Discrete Data
• Causality and Generalizability: Identifiability and Learning Methods
• Causal Inference on Multivariate and Mixed-Type Data
• Parallel ensemble methods for causal direction inference
• Entropic Causality and Greedy Minimum Entropy Coupling
• Information-Theoretic Approximation to Causal Models
• Gradient-based Causal Structure Learning with Normalizing Flow