Dichotomization invariant log-mean linear parameterization for discrete graphical models of marginal independence

TL;DR

This paper generalizes the log-mean linear parameterization for discrete graphical models to variables with multiple levels, enabling flexible representation of marginal independencies and reducing parameter complexity.

Contribution

It extends the log-mean linear parameterization to multi-level discrete variables and demonstrates its utility in modeling marginal independencies with potential applications in genetics.

Findings

Allows representation of marginal independencies with collapsed levels.

Reduces parameter count while preserving independence structure.

Provides insights into variable association structures.

Abstract

We extend the log-mean linear parameterization introduced by Roverato et al. (2013) for binary data to discrete variables with arbitrary number of levels, and show that also in this case it can be used to parameterize bi-directed graph models. Furthermore, we show that the log-mean linear parameterization allows one to simultaneously represent marginal independencies among variables and marginal independencies that only appear when certain levels are collapsed into a single one. We illustrate the application of this property by means of an example based on genetic association studies involving single-nucleotide polymorphisms. More generally, this feature provides a natural way to reduce the parameter count, while preserving the independence structure, by means of substantive constraints that give additional insight into the association structure of the variables.