Identification of Strong Edges in AMP Chain Graphs

TL;DR

This paper introduces a method to identify strong, invariant edges in essential AMP chain graphs, enabling better causal effect bounding when the true graph structure is uncertain.

Contribution

It develops a procedure to determine strong edges in essential AMP chain graphs, improving causal inference under graph uncertainty.

Findings

Identified criteria for strong edges in essential AMP chain graphs

Enabled bounding of causal effects with unknown true graph

Enhanced understanding of edge invariance in Markov equivalence classes

Abstract

The essential graph is a distinguished member of a Markov equivalence class of AMP chain graphs. However, the directed edges in the essential graph are not necessarily strong or invariant, i.e. they may not be shared by every member of the equivalence class. Likewise for the undirected edges. In this paper, we develop a procedure for identifying which edges in an essential graph are strong. We also show how this makes it possible to bound some causal effects when the true chain graph is unknown.