August 2022 Concentration of Markov chains indexed by trees
Christopher Shriver
Author Affiliations +
Ann. Inst. H. Poincaré Probab. Statist. 58(3): 1701-1711 (August 2022). DOI: 10.1214/21-AIHP1224

Abstract

An inequality of Marton (Ann. Probab. 24 (1996) 857–866) shows that the joint distribution of a Markov chain with uniformly contracting transition kernels exhibits concentration. We generalize this inequality to Markov chains indexed by trees.

Une inégalité de Marton (Ann. Probab. 24 (1996) 857–866) montre que la loi jointe d’une chaîne de Markov avec des noyaux de transition uniformément contractants présente un phénomène de concentration. Nous généralisons cette inégalité aux chaînes de Markov indexées par des arbres.

Funding Statement

This material is based upon work supported by the National Science Foundation under Grant No. DMS 1344970.

Acknowledgements

The author would like to thank Tim Austin and Georg Menz for many helpful discussions and feedback on earlier versions of the paper, and Aryeh Kontorovich and anonymous reviewers for providing feedback on preprints.

Citation

Download Citation

Christopher Shriver. "Concentration of Markov chains indexed by trees." Ann. Inst. H. Poincaré Probab. Statist. 58 (3) 1701 - 1711, August 2022. https://doi.org/10.1214/21-AIHP1224

Information

Received: 2 March 2020; Revised: 27 September 2021; Accepted: 15 October 2021; Published: August 2022
First available in Project Euclid: 14 July 2022

MathSciNet: MR4452649
zbMATH: 1492.60290
Digital Object Identifier: 10.1214/21-AIHP1224

Subjects:
Primary: 60K35

Keywords: concentration of measure , Ising model , Markov chains

Rights: Copyright © 2022 Association des Publications de l’Institut Henri Poincaré

Vol.58 • No. 3 • August 2022
Back to Top