• Corpus ID: 235303703

Locally associated graphical models and mixed convex exponential families

@inproceedings{Lauritzen2020LocallyAG,
  title={Locally associated graphical models and mixed convex exponential families},
  author={Steffen L. Lauritzen and Piotr Zwiernik},
  year={2020}
}
The notion of multivariate total positivity has proved to be useful in finance and psychology but may be too restrictive in other applications. In this paper we propose a concept of local association, where highly connected components in a graphical model are positively associated and study its properties. Our main motivation comes from gene expression data, where graphical models have become a popular exploratory tool. The models are instances of what we term mixed convex exponential families… 

Figures from this paper

Total positivity in multivariate extremes
TLDR
An estimator for the parameters of the Hüsler–Reiss distribution under EMTP2 is proposed as the solution of a convex optimization problem with Laplacian constraint and it is proved that this estimator is consistent and typically yields a sparse model with possibly non-decomposable extremal graphical structure.
Fast iterative proportional scaling for Gaussian graphical models
TLDR
This work presents a version of the algorithm which avoids repeated inversion of large matrices, resulting in increased speed, in particular when graphs are sparse, leading to increased speed in Gaussian graphical models.

References

SHOWING 1-10 OF 51 REFERENCES
Maximum likelihood estimation in Gaussian models under total positivity
We analyze the problem of maximum likelihood estimation for Gaussian distributions that are multivariate totally positive of order two (MTP2). By exploiting connections to phylogenetics and
The geometry of Gaussian double Markovian distributions
. Gaussian double Markovian models consist of covariance matrices constrained by a pair of graphs specifying zeros simultaneously in the matrix and its inverse. We study the semi-algebraic geometry
Covariance estimation with nonnegative partial correlations
TLDR
This work establishes that the Gaussian maximum likelihood estimator is both high-dimensionally consistent and minimax optimal in the symmetrized Stein loss and proves a negative result which shows that the sign-constraints can introduce substantial bias for estimating the top eigenvalue of the covariance matrix.
Dependence in elliptical partial correlation graphs
The Gaussian model is equipped with strong properties that facilitate studying and interpreting graphical models. Specifically it reduces conditional independence and the study of positive
Covariance Matrix Estimation under Total Positivity for Portfolio Selection*
TLDR
Based on stock-market data spanning over thirty years, it is shown that estimating the covariance matrix under MTP_2 outperforms previous state-of-the-art methods including shrinkage estimators and factor models.
Statistical Modelling by Exponential Families
This book is a readable, digestible introduction to exponential families, encompassing statistical models based on the most useful distributions in statistical theory, including the normal, gamma,
Total positivity in exponential families with application to binary variables
We study exponential families of distributions that are multivariate totally positive of order 2 (MTP2), show that these are convex exponential families, and derive conditions for existence of the
Statistical Properties of I-projections Within Exponential Families
La methode de I-projection, utilisee generalement en theorie de l'information, est etudiee pour les familles exponentielles d'un point de vue statistique, et une formule d'approximation pour la
...
...