PreprintarXiv licencearXiv
On the existence of a spectral gap for ergodic Markov chains
Abstract
Characterizations are established regarding the existence of an L2-spectral gap for ergodic (that is, irreducible, aperiodic, and positive) Markov chains on infinite state spaces. In particular, it is shown that the existence of an L2-spectral gap is entirely determined by the behavior of the spectrum at 1. These results are applied in the analysis of a family of Markov chains constructed to show that the ergodic composition of two geometrically ergodic Markov kernels need not be geometrically ergodic.
Related
- On the Information Bottleneck for Ergodic Finite-State Markov Chains: Stationary Approximation and Asymptotic Optimality
- Interpretable Social Network Churn Prediction via Ergodic Markov Chains
- Minimax estimation of the pseudo-spectral gap of continuous Markov chains, with applications to empirical concentration bounds
- Resolvent intertwining and spectral duality in Markov chains with geometric resetting
- Existence of ergodic invariant measures for 3D Navier-Stokes eqs with superlinear noise
BibTeX
@misc{mak2026existence,
title = {{On the existence of a spectral gap for ergodic Markov chains}},
author = {Xavier Mak and James P. Hobert},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.24808},
eprint = {2610.24808},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.24808},
}
RIS
TY - UNPB
TI - On the existence of a spectral gap for ergodic Markov chains
AU - Xavier Mak
AU - James P. Hobert
PY - 2026
DA - 2026/10/17
PB - arXiv
DO - 10.48550/arxiv.2610.24808
UR - https://arxiv.org/abs/2610.24808
AB - Characterizations are established regarding the existence of an L^2-spectral gap for ergodic (that is, irreducible, aperiodic, and positive) Markov chains on infinite state spaces. In particular, it is shown that the existence of an L^2-spectral gap is entirely determined by the behavior of the spectrum at 1. These results are applied in the analysis of a family of Markov chains constructed to show that the ergodic composition of two geometrically ergodic Markov kernels need not be geometrically ergodic.
ER -