Multivariate Generalized Gaussian Distribution: Convexity and Graphical Models

Abstract

We consider covariance estimation in the multivariate generalized Gaussian distribution (MGGD) and elliptically symmetric (ES) distribution. The maximum likelihood optimization associated with this problem is non-convex, yet it has been proved that its global solution can be often computed via simple fixed point iterations. Our first contribution is a new analysis of this likelihood based on geodesic convexity that requires weaker assumptions. Our second contribution is a generalized framework for structured covariance estimation under sparsity constraints. We show that the optimizations can be formulated as convex minimization as long the MGGD shape parameter is larger than half and the sparsity pattern is chordal. These include, for example, maximum likelihood estimation of banded inverse covariances in multivariate Laplace distributions, which are associated with time varying autoregressive processes.

References in corpus (3)

• Regularized estimation of large covariance matrices
• Parameter Estimation For Multivariate Generalized Gaussian Distributions
• Robust graphical modeling of gene networks using classical and alternative T-distributions

Cited by in corpus (24)

• Parameter Estimation For Multivariate Generalized Gaussian Distributions
• Conic geometric optimisation on the manifold of positive definite matrices
• An Overview of Robust Subspace Recovery
• Regularized $M$-estimators of scatter matrix
• Riemannian SVRG: Fast Stochastic Optimization on Riemannian Manifolds
• Power Systems Topology and State Estimation by Graph Blind Source Separation
• Tyler's Covariance Matrix Estimator in Elliptical Models with Convex Structure
• Performance Analysis of Tyler's Covariance Estimator
• First-order Methods for Geodesically Convex Optimization
• R-SPIDER: A Fast Riemannian Stochastic Optimization Algorithm with Curvature Independent Rate
• Simultaneous penalized M-estimation of covariance matrices using geodesically convex optimization
• Exponential ergodicity of mirror-Langevin diffusions
• Coupled regularized sample covariance matrix estimator for multiple classes
• Solving general elliptical mixture models through an approximate Wasserstein manifold
• Identifiability and optimal rates of convergence for parameters of multiple types in finite mixtures
• A universal framework for learning the elliptical mixture model
• Inference and Mixture Modeling with the Elliptical Gamma Distribution
• Shrinking the Sample Covariance Matrix using Convex Penalties on the Matrix-Log Transformation
• Marchenko-Pastur Law for Tyler's M-estimator
• Linear pooling of sample covariance matrices
• On the convergence of Maronna's $M$-estimators of scatter
• Geometrical and statistical properties of M-estimates of scatter on Grassmann manifolds
• Riemannian information gradient methods for the parameter estimation of ECD: Some applications in image processing
• Mixtures of Multivariate Power Exponential Distributions