Open Access
May 2008 Gibbs Sampling, Exponential Families and Orthogonal Polynomials
Persi Diaconis, Kshitij Khare, Laurent Saloff-Coste
Statist. Sci. 23(2): 151-178 (May 2008). DOI: 10.1214/07-STS252

Abstract

We give families of examples where sharp rates of convergence to stationarity of the widely used Gibbs sampler are available. The examples involve standard exponential families and their conjugate priors. In each case, the transition operator is explicitly diagonalizable with classical orthogonal polynomials as eigenfunctions.

Citation

Download Citation

Persi Diaconis. Kshitij Khare. Laurent Saloff-Coste. "Gibbs Sampling, Exponential Families and Orthogonal Polynomials." Statist. Sci. 23 (2) 151 - 178, May 2008. https://doi.org/10.1214/07-STS252

Information

Published: May 2008
First available in Project Euclid: 21 August 2008

zbMATH: 1327.62058
MathSciNet: MR2446500
Digital Object Identifier: 10.1214/07-STS252

Keywords: conjugate priors , exponential families , Gibbs sampler , location families , orthogonal polynomials , running time analyses , Singular value decomposition

Rights: Copyright © 2008 Institute of Mathematical Statistics

Vol.23 • No. 2 • May 2008
Back to Top