Causal Discovery with Unobserved Confounding and non-Gaussian Data

by   Y. Samuel Wang, et al.

We consider the problem of recovering causal structure from multivariate observational data. We assume that the data arise from a linear structural equation model (SEM) in which the idiosyncratic errors are allowed to be dependent in order to capture possible latent confounding. Each SEM can be represented by a graph where vertices represent observed variables, directed edges represent direct causal effects, and bidirected edges represent dependence among error terms. Specifically, we assume that the true model corresponds to a bow-free acyclic path diagram, i.e., a graph that has at most one edge between any pair of nodes and is acyclic in the directed part. We show that when the errors are non-Gaussian, the exact causal structure encoded by such a graph, and not merely an equivalence class, can be consistently recovered from observational data. The Bow-free Acylic Non-Gaussian (BANG) method we propose for this purpose uses estimates of suitable moments, but, in contrast to previous results, does not require specifying the number of latent variables a priori. We illustrate the effectiveness of BANG in simulations and an application to an ecology data set.



There are no comments yet.


page 1

page 2

page 3

page 4


Bayesian network structure learning with causal effects in the presence of latent variables

Latent variables may lead to spurious relationships that can be misinter...

Learning Linear Non-Gaussian Causal Models in the Presence of Latent Variables

We consider the problem of learning causal models from observational dat...

Structure Learning for Cyclic Linear Causal Models

We consider the problem of structure learning for linear causal models b...

Causal Model Analysis using Collider v-structure with Negative Percentage Mapping

A major problem of causal inference is the arrangement of dependent node...

Half-Trek Criterion for Identifiability of Latent Variable Models

We consider linear structural equation models with latent variables and ...

High-Dimensional Causal Discovery Under non-Gaussianity

We consider data from graphical models based on a recursive system of li...

High-Dimensional Discovery Under non-Gaussianity

We consider data from graphical models based on a recursive system of li...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.