Random walks in Euclidean space

From MaRDI portal
(Redirected from Publication:482912)



Abstract: Consider a sequence of independent random isometries of Euclidean space with a previously fixed probability law. Apply these isometries successively to the origin and consider the sequence of random points that we obtain this way. We prove a local limit theorem under a suitable moment condition and a necessary non-degeneracy condition. Under stronger hypothesis, we prove a limit theorem on a wide range of scales: between e^(-cl^(1/4)) and l^(1/2), where l is the number of steps.


Let \(X_1, X_2,\dots\) be independent identically distributed random isometries of the Euclidean space \(R^d\). Let \(x_0\in R^d\) be any point, and consider the sequence of points \[ Y=x_0, \dots, Y_e(X_{l-1}(\dots (x_0))), \dots. \] The author's main objective is to obtain a local limit theorem in the following sense. Under some conditions on \(Y_0\) and \(X\) there are \(v_0\in \mathbb{R}^d\) and \(c>0\) depending only on the distribution of \(X_j\) such that \[ \lim_{l\to0}l^{d/2}E[f(Y_0-lv_0)]=c\int f(r)\,dx, \] where \(f\) is any continuous and compactly supported function.











This page was built for publication: Random walks in Euclidean space

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q482912)