Elsevier

Discrete Applied Mathematics

Volume 244, 31 July 2018, Pages 170-185
Discrete Applied Mathematics

Counting Markov equivalence classes for DAG models on trees

https://doi.org/10.1016/j.dam.2018.03.015Get rights and content
Under an Elsevier user license
open archive

Abstract

DAG models are statistical models satisfying a collection of conditional independence relations encoded by the nonedges of a directed acyclic graph (DAG) G. Such models are used to model complex cause–effect systems across a variety of research fields. From observational data alone, a DAG model G is only recoverable up to Markov equivalence. Combinatorially, two DAGs are Markov equivalent if and only if they have the same underlying undirected graph (i.e., skeleton) and the same set of the induced subDAGs ijk, known as immoralities. Hence it is of interest to study the number and size of Markov equivalence classes (MECs). In a recent paper, we introduced a pair of generating functions that enumerate the number of MECs on a fixed skeleton by number of immoralities and by class size, and we studied the complexity of computing these functions. In this paper, we lay the foundation for studying these generating functions by analyzing their structure for trees and other closely related graphs. We describe these polynomials for some well-studied families of graphs including paths, stars, cycles, spider graphs, caterpillars, and balanced binary trees. In doing so, we recover connections to independence polynomials, and extend some classical identities that hold for Fibonacci numbers. We also provide tight lower and upper bounds for the number and size of MECs on any tree. Finally, we use computational methods to show that the number and distribution of high degree nodes in a triangle-free graph dictate the number and size of MECs.

Keywords

DAG model
Bayesian network
Markov equivalence class
Markov equivalence
Trees
Immoralities

Cited by (0)