Generating random Gaussian graphical models

09/03/2019
by   Irene Córdoba, et al.
0

Structure learning methods for covariance and concentration graphs are often validated on synthetic models, usually obtained by randomly generating: (i) an undirected graph, and (ii) a compatible symmetric positive definite (SPD) matrix. In order to ensure positive definiteness in (ii), a dominant diagonal is usually imposed. In this work we investigate different methods to generate random symmetric positive definite matrices with undirected graphical constraints. We show that if the graph is chordal is possible to sample uniformly from the set of correlation matrices compatible with the graph, while for general undirected graphs we rely on a partial orthogonalization method.

READ FULL TEXT
research
07/09/2018

A partial orthogonalization method for simulating covariance and concentration graph matrices

Structure learning methods for covariance and concentration graphs are o...
research
06/12/2019

Learning High-Dimensional Gaussian Graphical Models under Total Positivity without Tuning Parameters

We consider the problem of estimating an undirected Gaussian graphical m...
research
12/03/2020

Coloured Graphical Models and their Symmetries

Coloured graphical models are Gaussian statistical models determined by ...
research
06/17/2021

Identifiability of AMP chain graph models

We study identifiability of Andersson-Madigan-Perlman (AMP) chain graph ...
research
10/21/2022

Learning Graphical Factor Models with Riemannian Optimization

Graphical models and factor analysis are well-established tools in multi...
research
07/12/2019

Path Weights in Concentration Graphs

A graphical model provides a compact and efficient representation of the...
research
09/23/2012

Gemini: Graph estimation with matrix variate normal instances

Undirected graphs can be used to describe matrix variate distributions. ...

Please sign up or login with your details

Forgot password? Click here to reset