On convergence rates of finite memory estimators of infinite memory processes

Imre Csiszár, Zsolt Talata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Stationary ergodic processes with finite alphabets are approximated by finite memory processes based on an n-length realization of the process. Under the assumptions of summable continuity rate and non-nullness, a rate of convergence in d̄-distance is obtained, with explicit constants. Asymptotically, as n → ∞, the result is near the optimum.

Original languageEnglish
Title of host publicationIEEE Information Theory Workshop 2010, ITW 2010
DOIs
Publication statusPublished - Jul 27 2010
EventIEEE Information Theory Workshop 2010, ITW 2010 - Cairo, Egypt
Duration: Jan 6 2010Jan 8 2010

Publication series

NameIEEE Information Theory Workshop 2010, ITW 2010

Other

OtherIEEE Information Theory Workshop 2010, ITW 2010
CountryEgypt
CityCairo
Period1/6/101/8/10

    Fingerprint

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Theoretical Computer Science

Cite this

Csiszár, I., & Talata, Z. (2010). On convergence rates of finite memory estimators of infinite memory processes. In IEEE Information Theory Workshop 2010, ITW 2010 [5503181] (IEEE Information Theory Workshop 2010, ITW 2010). https://doi.org/10.1109/ITWKSPS.2010.5503181