Non-asymptotic Estimates for Markov Transition Matrices via Spectral Gap Methods

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, De, Li, Xiangyuan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917063153418240
author Huang, De
Li, Xiangyuan
author_facet Huang, De
Li, Xiangyuan
contents We establish non-asymptotic error bounds for the classical Maximal Likelihood Estimation of the transition matrix of a given Markov chain. Meanwhile, in the reversible case, we propose a new reversibility-preserving online Symmetric Counting Estimation of the transition matrix with non-asymptotic deviation bounds. Our analysis is based on a convergence study of certain Markov chains on the length-2 path spaces induced by the original Markov chain.
format Preprint
id arxiv_https___arxiv_org_abs_2408_05963
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Non-asymptotic Estimates for Markov Transition Matrices via Spectral Gap Methods
Huang, De
Li, Xiangyuan
Statistics Theory
Probability
60J10, 37A25, 62M05
We establish non-asymptotic error bounds for the classical Maximal Likelihood Estimation of the transition matrix of a given Markov chain. Meanwhile, in the reversible case, we propose a new reversibility-preserving online Symmetric Counting Estimation of the transition matrix with non-asymptotic deviation bounds. Our analysis is based on a convergence study of certain Markov chains on the length-2 path spaces induced by the original Markov chain.
title Non-asymptotic Estimates for Markov Transition Matrices via Spectral Gap Methods
topic Statistics Theory
Probability
60J10, 37A25, 62M05
url https://arxiv.org/abs/2408.05963