Large Deviations for a random walk in random environment

Greven A, den Hollander F (1994)


Publication Language: English

Publication Type: Journal article, Original article

Publication year: 1994

Journal

Publisher: Institute of Mathematical Statistics (IMS)

Book Volume: 22

Pages Range: 1381-1428

Journal Issue: 3

URI: https://projecteuclid.org/euclid.aop/1176988607

DOI: 10.1214/aop/1176988607

Abstract

Let ω=(px)x∈Z" role="presentation">ω=(px)x∈ℤ be an i.i.d. collection of (0, 1)-valued random variables. Given ω" role="presentation">ω, let (Xn)n≥0" role="presentation">(Xn)n≥0 be the Markov chain on Z" role="presentation">ℤ defined by X0=0" role="presentation">X0=0 and Xn+1=Xn+1(resp.Xn−1)" role="presentation">Xn+1=Xn+1(resp.Xn−1) with probability pXn(resp⁡.1−pXn)" role="presentation">pXn(resp.1−pXn). It is shown that Xn/n" role="presentation">Xn/n satisfies a large deviation principle with a continuous rate function, that is, limn→∞1nlog⁡Pω(Xn=⌊θnn⌋)=−I(θ)ω−a.s.forθn→∈[−1,1]." role="presentation">limn→∞1nlogPω(Xn=⌊θnn⌋)=−I(θ)ω−a.s.forθn→∈[−1,1]. First, we derive a representation of the rate function I" role="presentation">I in terms of a variational problem. Second, we solve the latter explicitly in terms of random continued fractions. This leads to a classification and qualitative description of the shape of I" role="presentation">I. In the recurrent case I" role="presentation">I is nonanalytic at θ=0" role="presentation">θ=0. In the transient case I" role="presentation">I is nonanalytic at θ=−θc,0,θc" role="presentation">θ=−θc,0,θc for some θc≥0" role="presentation">θc≥0, with linear pieces in between.

Authors with CRIS profile

Involved external institutions

How to cite

APA:

Greven, A., & den Hollander, F. (1994). Large Deviations for a random walk in random environment. Annals of Probability, 22(3), 1381-1428. https://doi.org/10.1214/aop/1176988607

MLA:

Greven, Andreas, and Frank den Hollander. "Large Deviations for a random walk in random environment." Annals of Probability 22.3 (1994): 1381-1428.

BibTeX: Download