Bayesian Graphical Models for Multivariate Functional Data

TL;DR

This paper develops a Bayesian framework for graphical models of multivariate functional data, extending Markov properties to functions and proposing new priors for covariance kernels, with applications in brain activity analysis.

Contribution

It introduces a novel notion of conditional independence for functions and constructs a Bayesian inference framework for Gaussian process graphical models with new hyper-inverse-Wishart-process priors.

Findings

Effective stochastic algorithms for posterior inference.

Successful application to brain activity data.

Framework extends graphical models to functional data.

Abstract

Graphical models express conditional independence relationships among variables. Although methods for vector-valued data are well established, functional data graphical models remain underdeveloped. We introduce a notion of conditional independence between random functions, and construct a framework for Bayesian inference of undirected, decomposable graphs in the multivariate functional data context. This framework is based on extending Markov distributions and hyper Markov laws from random variables to random processes, providing a principled alternative to naive application of multivariate methods to discretized functional data. Markov properties facilitate the composition of likelihoods and priors according to the decomposition of a graph. Our focus is on Gaussian process graphical models using orthogonal basis expansions. We propose a hyper-inverse-Wishart-process prior for the covariance kernels of the infinite coefficient sequences of the basis expansion, establish existence, uniqueness, strong hyper Markov property, and conjugacy. Stochastic search Markov chain Monte Carlo algorithms are developed for posterior inference, assessed through simulations, and applied to a study of brain activity and alcoholism.