litlas
Log inCreate account

A Linear Non-Gaussian Acyclic Model for Causal Discovery

Authors: Shohei Shimizu, Patrik O. Hoyer, Aapo Hyvärinen, Antti J. Kerminen - Journal of Machine Learning Research, JMLR 2006 cited by 1,177

In recent years, several methods have been proposed for the discovery of causal structure from non-experimental data. Such methods make various assumptions on the data generating process to facilitate its identification from purely observational data. Continuing this line of research, we show how to discover the complete causal structure of continuous-valued data, under the assumptions that (a) the data generating process is linear, (b) there are no unobserved confounders, and (c) disturbance variables have non-Gaussian distributions of non-zero variances. The solution relies on the use of the statistical method known as independent component analysis, and does not require any pre-specified time-ordering of the variables. We provide a complete Matlab package for performing this LiNGAM analysis (short for Linear Non-Gaussian Acyclic Model), and demonstrate the effectiveness of the method using artificially generated data and real-world data.

A Sharper BonferroniProcedure for Multiple…A Sharper Bonferroni Procedure for Multiple Tests of SignificanceStructural Equationswith Latent Variables.Structural Equations with Latent Variables.An Introduction to theBootstrapAn Introduction to the BootstrapControlling the FalseDiscovery Rate: A…Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple TestingEquivariant adaptivesource separationEquivariant adaptive source separationFast and robustfixed-point algorithms…Fast and robust fixed-point algorithms for independent component analysisCausality: Models,Reasoning and InferenceCausality: Models, Reasoning and InferenceIndependent ComponentAnalysisIndependent Component AnalysisOne-unit contrastfunctions for…One-unit contrast functions for independent component analysis: a statistical analysisTesting Significance ofMixing and Demixing…Testing Significance of Mixing and Demixing Coefficients in ICAPerformance analysis ofthe FastICA algorithm…Performance analysis of the FastICA algorithm and Cramér-rao bounds for linear independent component analysisUse of non-normality instructural equation…Use of non-normality in structural equation modeling: Application to direction of causationParceLiNGAM: A CausalOrdering Method Robust…ParceLiNGAM: A Causal Ordering Method Robust Against Latent ConfoundersDistinguishing betweencause and effectDistinguishing between cause and effectOn the Identifiabilityof the Post-Nonlinear…On the Identifiability of the Post-Nonlinear Causal ModelIdentifiability ofCausal Graphs using…Identifiability of Causal Graphs using Functional ModelsSparse LinearIdentifiable…Sparse Linear Identifiable Multivariate ModelingCausal discovery withcontinuous additive…Causal discovery with continuous additive noise modelsBayesian networks forfMRI: A primerBayesian networks for fMRI: A primerScore-based causallearning in additive…Score-based causal learning in additive noise modelsDirection of effects inmediation analysis.Direction of effects in mediation analysis.An EfficientEntropy-Based Causal…An Efficient Entropy-Based Causal Discovery Method for Linear Structural Equation Models With IID Noise VariablesBayesian Estimation ofCausal Direction in…Bayesian Estimation of Causal Direction in Acyclic Structural Equation Models with Individual-specific Confounder Variables and Non-Gaussian DistributionsLearning latent causalgraphs via mixture…Learning latent causal graphs via mixture oraclesA Linear Non-GaussianAcyclic Model for Causa…A Linear Non-Gaussian Acyclic Model for Causal DiscoveryEarlier referencesFocus paperCiting papersOlderNewer

Click a node to pin it, click the empty canvas to go back to this paper, or hover to preview. Open a node’s page from its title.