Closures of exponential families

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.

Published at http://dx.doi.org/10.1214/009117904000000766 in the Annals of Probability (http://www.imstat.org/aop/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (10)

• Maximum likelihood estimation in log-linear models
• Support Sets in Exponential Families and Oriented Matroid Theory
• Quantum Convex Support
• Algebraic statistical models
• Asymptotic quantization of exponential random graphs
• A note on the border of an exponential family
• Pre-images of extreme points of the numerical range, and applications
• Computational information geometry: theory and practice
• Logistic regression geometry
• Maximizing the Bregman divergence from a Bregman family