Decomposable context-specific models

TL;DR

This paper introduces decomposable context-specific models derived from staged tree models, providing algebraic and combinatorial characterizations, and demonstrating their properties generalize those of decomposable graphical models.

Contribution

It defines a new class of decomposable context-specific models, characterizes their independence relations algebraically, and extends properties of decomposable graphical models to this setting.

Findings

Algebraic and combinatorial characterization of independence relations

Moralization does not alter implied independence relations

Generalization of properties from decomposable graphical models

Abstract

We introduce a family of discrete context-specific models, which we call decomposable. We construct this family from the subclass of staged tree models known as CStree models. We give an algebraic and combinatorial characterization of all context-specific independence relations that hold in a decomposable context-specific model, which yields a Markov basis. We prove that the moralization operation applied to the graphical representation of a context-specific model does not affect the implied independence relations, thus affirming that these models are algebraically described by a finite collection of decomposable graphical models. More generally, we establish that several algebraic, combinatorial, and geometric properties of decomposable context-specific models generalize those of decomposable graphical models to the context-specific setting.