Causal Discovery with Unobserved Confounding and non-Gaussian Data

07/21/2020
by   Y. Samuel Wang, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/29/2020

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

Latent variables may lead to spurious relationships that can be misinter...
research
08/11/2019

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

We consider the problem of learning causal models from observational dat...
research
06/10/2020

Structure Learning for Cyclic Linear Causal Models

We consider the problem of structure learning for linear causal models b...
research
09/16/2015

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

A major problem of causal inference is the arrangement of dependent node...
research
01/12/2022

Half-Trek Criterion for Identifiability of Latent Variable Models

We consider linear structural equation models with latent variables and ...
research
03/29/2018

High-Dimensional Causal Discovery Under non-Gaussianity

We consider data from graphical models based on a recursive system of li...
research
03/29/2018

High-Dimensional Discovery Under non-Gaussianity

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

Please sign up or login with your details

Forgot password? Click here to reset