Bayesian hypothesis testing for Gaussian graphical models: Conditional independence and order constraints

Donald R. Williams*, Joris Mulder

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)

Abstract

Gaussian graphical models (GGM; partial correlation networks) have become increasingly popular in the social and behavioral sciences for studying conditional (in)dependencies between variables. In this work, we introduce exploratory and confirmatory Bayesian tests for partial correlations. For the former, we first extend the customary GGM formulation that focuses on conditional dependence to also consider the null hypothesis of conditional independence for each partial correlation. Here a novel testing strategy is introduced that can provide evidence for a null, negative, or positive effect. We then introduce a test for hypotheses with order constraints on partial correlations. This allows for testing theoretical and clinical expectations in GGMs. The novel matrix- prior distribution is described that provides increased flexibility in specification compared to the Wishart prior. The methods are applied to PTSD symptoms. In several applications, we demonstrate how the exploratory and confirmatory approaches can work in tandem: hypotheses are formulated from an initial analysis and then tested in an independent dataset. The methodology is implemented in the R package BGGM.
Original languageEnglish
Article number102441
JournalJournal of Mathematical Psychology
Volume99
DOIs
Publication statusPublished - 2020

Keywords

  • Bayes factor
  • Gaussian graphical model
  • Matrix-F prior
  • Order-constraints
  • Partial correlation

Fingerprint Dive into the research topics of 'Bayesian hypothesis testing for Gaussian graphical models: Conditional independence and order constraints'. Together they form a unique fingerprint.

Cite this