Third-Order Moment Varieties of Linear Non-Gaussian Graphical Models

TL;DR

This paper explores the algebraic structure of third-order moments in linear non-Gaussian graphical models, revealing toric ideals for polytrees and providing explicit constructions for their defining relations.

Contribution

It introduces a novel algebraic framework for understanding third-order moments in non-Gaussian models, including explicit trek-matrix constructions and ideal descriptions for models with hidden variables.

Findings

Third-order moments form toric ideals in polytrees.

Explicit trek-matrices generate the model's ideal.

Descriptions of polytopes and ideals for models with hidden variables.

Abstract

In this paper we study linear non-Gaussian graphical models from the perspective of algebraic statistics. These are acyclic causal models in which each variable is a linear combination of its direct causes and independent noise. The underlying directed causal graph can be identified uniquely via the set of second and third order moments of all random vectors that lie in the corresponding model. Our focus is on finding the algebraic relations among these moments for a given graph. We show that when the graph is a polytree these relations form a toric ideal. We construct explicit trek-matrices associated to 2-treks and 3-treks in the graph. Their entries are covariances and third order moments and their 2-minors define our model set-theoretically. Furthermore, we prove that their 2-minors also generate the vanishing ideal of the model. Finally, we describe the polytopes of third order moments and the ideals for models with hidden variables.