Closures of exponential families

DOI10.1214/009117904000000766MaRDI QIDQ1775441zbMATH OpenOpenAlexFDO

Authors Imre Csiszár, František Matúš

Publication date 3 May 2005

Published in The Annals of Probability (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0503653

zbMATH Keywords

accessible faceclosure propertiesconvex coreconvex supportexponential family of probability measuresextensioninformation divergencereverse information divergencevariation distanceweak convergence

Mathematics Subject Classification ID

Convex sets in (n) dimensions (including convex hypersurfaces) (52A20) Probabilistic measure theory (60A10) Convergence of probability measures (60B10) Statistical aspects of information-theoretic topics (62B10)

Abstract: The variation distance closure of an exponential family with a convex set of canonical parameters is described, assuming no regularity conditions. The tools are the concepts of convex core of a measure and extension of an exponential family, introduced previously by the authors, and a new concept of accessible faces of a convex set. Two other closures related to the information divergence are also characterized.

Review text

The paper considers exponential families E of probability measures (PMs) having both discrete and continuous components as a convex set of canonical parameters. The main purpose of this paper is to investigate, in the general case when all common regularity conditions are absent, the properties of three important closures for exponential families E: the variation (distance) closure (V-closure), the information (divergence) closure (I-closure), and reverse I-closure (rI-closure). The essential tools used are the newly defined notions of the convex core of a measure, the accessible faces of a convex set, and the extension of an exponential family E. The paper is organized as follows: Section 2 introduces the necessary notions and preliminary results, Section 3 develops in detail the main properties of the three closures of E families, while Section 4 is devoted to the weak convergence in exponential families. The final Sections 5 and 6 expose comprehensive proofs of previously enclosed technical results. The investigated closures are of major relevance in large deviation theory, for a generalized maximum likelihood estimate, in canonical statistics etc.

Recommendations

• Generalized maximum likelihood estimates for exponential families
• On limiting towards the boundaries of exponential families.
• Support sets in exponential families and oriented matroid theory
• Optimally approximating exponential families
• On natural parameter space of exponential family

Cites work

• Convex Analysis
• Convex cores of measures on R d
• scientific article; zbMATH DE number 1667417 (Why is no real title available?)
• scientific article; zbMATH DE number 3761167 (Why is no real title available?)
• scientific article; zbMATH DE number 44577 (Why is no real title available?)
• scientific article; zbMATH DE number 3602484 (Why is no real title available?)
• Information projections revisited
• Lectures on natural exponential families and their variance functions
• On Information Closures of Exponential Families: A Counterexample
• Sanov property, generalized I-projection and a conditional limit theorem

Cited in

(17)

• On the geometry of discrete exponential families with application to exponential random graph models
• Computationally efficient likelihood inference in exponential families when the maximum likelihood estimator does not exist
• Generalized maximum likelihood estimates for exponential families
• Computing with Fisher geodesics and extended exponential families
• On limiting towards the boundaries of exponential families.
• On Information Closures of Exponential Families: A Counterexample
• scientific article; zbMATH DE number 63439 (Why is no real title available?)
• Maximum likelihood estimation in log-linear models
• Optimally approximating exponential families
• Optimality conditions for maximizers of the information divergence from an exponent family
• Support sets in exponential families and oriented matroid theory
• Maximizing the Bregman divergence from a Bregman family
• Geometry and applied statistics
• When are first-order asymptotics adequate? A diagnostic
• Geodesic non-completeness of the truncated normal family
• Quantum convex support
• Joint numerical ranges of three Hermitian 4 4 matrices

This page was built for publication: Closures of exponential families

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1775441)