Large deviations application to Billingsley's example

DOI10.1080/07474946.2010.487429MaRDI QIDQ4931848zbMATH OpenOpenAlexFDO

Authors R. Liptser

Publication date 1 October 2010

Published in Sequential Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0905.4334

zbMATH Keywords

empirical distributionlarge deviations principlestopping time

Mathematics Subject Classification ID

Large deviations (60F10) Functional limit theorems; invariance principles (60F17) Martingales with continuous parameter (60G44) Continuous-time Markov processes on discrete state spaces (60J27)

Abstract: We consider a classical model related to an empirical distribution function $F_n(t)=frac1nsu{m}_{k=1}^n{I}_{x{i}_{k}let}$ of $(x{i}_k{)}_{ige1}$ -- i.i.d. sequence of random variables, supported on the interval $[0,1]$, with continuous distribution function $F(t)=mathsfP(x{i}_{1}let)$. Applying ``Stopping Time Techniques, we give a proof of Kolmogorov's exponential bound ${\displaystyle }$ mathsf{P}�ig(sup_{tin[0,1]}|F_n(t)-F(t)|ge varepsilon�ig)le ext{const.}e^{-ndelta_varepsilon} ${\displaystyle }$ conjectured by Kolmogorov in 1943. Using this bound we establish a best possible logarithmic asymptotic of ${\displaystyle }$ mathsf{P}�ig(sup_{tin[0,1]}n^alpha|F_n(t)-F(t)|ge varepsilon�ig) ${\displaystyle }$ with rate $frac1n^{1-2alpha}$ slower than $frac1n$ for any $$.

Recommendations

• Moderately large deviations and expansions of large deviations for some functionals of weighted empirical processes
• scientific article; zbMATH DE number 4124809
• scientific article; zbMATH DE number 4064154
• Improved Kolmogorov inequalities for the binomial distribution
• The principle of large deviations for empirical processes

Cites work

• scientific article; zbMATH DE number 3181534 (Why is no real title available?)
• scientific article; zbMATH DE number 44889 (Why is no real title available?)
• scientific article; zbMATH DE number 3274494 (Why is no real title available?)
• Large deviations, moderate deviations and LIL for empirical processes
• The method of stochastic exponentials for large deviations

Cited in

(1)

• Observing a Preferred Distribution

This page was built for publication: Large deviations application to Billingsley's example

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4931848)