Graphical models with marginals in the exponential family
Preprint
Abstract
Graphical models encode conditional independence statements of a multivariate distribution via a graph. Traditionally, the marginal distributions in a graphical model are assumed to be Gaussian. In this paper, we propose a three-level hierarchical model that functions as the hyper-Markov law that enables a graphical model with marginals in the exponential family with quadratic variance function. Inference on the model parameters is made using a Bayesian approach. We perform a simulation study and real data analyses to illustrate the usefulness of our models.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Dependence Models via Hierarchical Structures2025
- Bayesian regression with spatiotemporal varying coefficientsBiometrical Journal, 2020
- Bayesian Regularization for Graphical Models With Unequal ShrinkageJournal of the American Statistical Association, 2018
- Bayesian Inference for Gaussian Graphical Models Beyond Decomposable GraphsJournal of the Royal Statistical Society Series B: Statistical Methodology, 2018
- Asymptotically Normal and Efficient Estimation of Covariate-Adjusted Gaussian Graphical ModelJournal of the American Statistical Association, 2016
- Efficient Bayesian inference for Gaussian copula regression modelsBiometrika, 2006
- Decomposable graphical Gaussian model determinationBiometrika, 1999
- Bayesian Computation Via the Gibbs Sampler and Related Markov Chain Monte Carlo MethodsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1993
- Natural Exponential Families with Quadratic Variance FunctionsThe Annals of Statistics, 1982
- Conjugate Priors for Exponential FamiliesThe Annals of Statistics, 1979