Supplementary material for Markov equivalence for ancestral graphs

TL;DR

This paper discusses the complexity of determining Markov equivalence in ancestral graphs, showing that existing criteria can involve exponentially many features, which impacts computational feasibility.

Contribution

It demonstrates that the criterion for Markov equivalence by Zhao et al. (2005) can involve an exponential number of graph features, highlighting computational challenges.

Findings

Markov equivalence criteria can be exponentially complex

Existing criteria may be computationally infeasible for large graphs

Highlights need for more efficient methods

Abstract

We prove that the criterion for Markov equivalence provided by Zhao et al. (2005) may involve a set of features of a graph that is exponential in the number of vertices.