scientific article; zbMATH DE number 3953933

From MaRDI portal
Publication:3723401





Consider a random walk \(S_ n=X_ 1,...,X_ n,\) \(n\geq 1\), generated by a sequence of i.i.d. random variables \(X_ n\), \(n\geq 1\), with values in a locally compact group G. The asymptotic behaviour of \(S_ n\) as \(n\to \infty\) depends on the properties of the distribution of the \(X_ n\) and those of G. An important part of that behaviour is played by the behaviour of the entropy \(H(S_ n)\) of \(S_ n\) as \(n\to \infty.\) In many cases, \(H(S_ n)\sim hn+c \log n\) as \(n\to \infty\). This formula has a remarkable probabilistic interpretation. The first term hn, where h is a constant called the asymptotic entropy of the random walk, is connected with the law of large numbers; the case \(h=0\) is typical of random walks for which the only harmonic functions are the constants. The second term c log n is connected with the central limit theorem. The aim of this paper is to make precise the above considerations. In the reviewer's opinion this is an important paper which, moreover, is written in a very readable manner.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3723401)