Published on Fri Mar 29 2013

ParceLiNGAM: A causal ordering method robust against latent confounders

Tatsuya Tashiro, Shohei Shimizu, Aapo Hyvarinen, Takashi Washio

We consider learning a causal ordering of variables in a linear non-Gaussian acyclic model called LiNGAM. The key idea is to detect latent confounders. We demonstrate the effectiveness of our method using artificial data and simulated brain imaging data.

0
0
0
Abstract

We consider learning a causal ordering of variables in a linear non-Gaussian acyclic model called LiNGAM. Several existing methods have been shown to consistently estimate a causal ordering assuming that all the model assumptions are correct. But, the estimation results could be distorted if some assumptions actually are violated. In this paper, we propose a new algorithm for learning causal orders that is robust against one typical violation of the model assumptions: latent confounders. The key idea is to detect latent confounders by testing independence between estimated external influences and find subsets (parcels) that include variables that are not affected by latent confounders. We demonstrate the effectiveness of our method using artificial data and simulated brain imaging data.

Mon Apr 09 2012
Machine Learning
Estimation of causal orders in a linear non-Gaussian acyclic model: a method robust against latent confounders
We consider to learn a causal ordering of variables in a linear non-Gaussian acyclic model called LiNGAM. We demonstrate the effectiveness of our method using artificial data.
0
0
0
Fri Apr 19 2019
Machine Learning
Causal Discovery with General Non-Linear Relationships Using Non-Linear ICA
We consider the problem of inferring causal relationships between two or more passively observed variables. The majority of current methods assume a linear causal relationship. We show rigorously that such non-linear ICA can be used to infer the causal direction via a series of independence tests.
0
0
0
Thu Jan 13 2011
Machine Learning
DirectLiNGAM: A direct method for learning a linear non-Gaussian structural equation model
Structural equation models and Bayesian networks have been widely used to analyze causal relations between continuous variables. In such frameworks, linear acyclic models are typically used to model the data-generating process. However, existing estimation methods are based on iterative search algorithms.
0
0
0
Mon Jan 13 2020
Machine Learning
Causal discovery of linear non-Gaussian acyclic models in the presence of latent confounders
Causal discovery from data affected by latent confounders is an important and difficult challenge. Causal functional model-based approaches have not been used to present variables whose relationships are affected. This paper proposes a method called repetitive causalisivelydiscovery to discover the causal structure of observed variables.
0
0
0
Fri Jun 04 2021
Machine Learning
Discovery of Causal Additive Models in the Presence of Unobserved Variables
Causal discovery from data affected by unobserved variables is an important but difficult problem to solve. Causal additive models exhibit structural equations that are additive in the variables and error terms. We propose a method to identify all the causal relationships that are theoretically theoreticallypossible to identify without being biased.
2
0
0
Fri Sep 15 2017
Machine Learning
Learning Functional Causal Models with Generative Neural Networks
Causal Generative Neural Networks (CGNN) leverages the power of neural networks to learn a generative model of the joint distribution of the observed variables. CGNN is applied to cause-effect inference, identifying the best causal hypothesis and identifying conditional independences.
0
0
0