Annealed deviations of random walk in random scenery (Q868050)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Annealed deviations of random walk in random scenery
scientific article

    Statements

    Annealed deviations of random walk in random scenery (English)
    0 references
    0 references
    0 references
    0 references
    19 February 2007
    0 references
    Let \((Z_n)_{n=1,2,\dots}\) be a \(d\)-dimensional random walk in random scenery, i.e., \(Z_n=\sum_{k=0}^{n-1} Y(S_k)\) with \((S_k)_{k=0,1,\dots}\) a random walk in \(\mathbb{Z}^d\) and \((Y(z))_{z \in \mathbb{Z}^d}\) an i.i.d. scenery, independent of the walk. The walker's step has mean zero and some finite exponential moments. The speed and the rate of \(\log P(n^{-1}Z_n >b_n)\) are identified for various choices of sequences \((b_n)_{n=1,2,\dots}\). For example, the so-called very large deviations are defined by requirements \(1 \ll b_n\ll n^{1/q}\) for some \(q>0\). Depending on \((b_n)_n\) and the upper tails of the scenery, the authors identify different regimes for the speed of \(\log P(n^{-1}Z_n >b_n)\) and different variational formulas for the rate functions. The considered problem has been previously addressed in the work of \textit{A. Asselah} and \textit{F. Castell} [Probab. Theory Relat. Fields 126, No.~4, 497--527 (2003; Zbl 1043.60018)]. The main novelty of the paper is the study of arbitrary sceneries unbounded to \(+\infty\) and general scale functions \(b_n \gg 1\) in the discrete setting. The results of the article have interesting connections to large deviation properties of self-intersections of the walk, which have been studied by \textit{X. Chen} [Ann. Probab. 32, No.~4, 3248--3300 (2004; Zbl 1067.60071)].
    0 references
    0 references
    0 references
    0 references
    0 references
    random walk in random scenery
    0 references
    local time
    0 references
    large deviations
    0 references
    variational formulas
    0 references
    0 references
    0 references