Laws of large numbers for hypergroups on \({\mathbb{R}}_+\) (Q1103938)

Let * be a convolution operation on the halfline and suppose that the Lebesgue density A of its Haar measure satisfies the convexity conditions of \textit{H. Chébli} [Théorie du Potentiel, Analyse Harmon.; Journées Soc. Math. France, Strasbourg 1973, Lecture Notes Math. 404, 35-59 (1974; Zbl 0304.43021)]. If \((S_ n:n\geq 0)\) is a random walk on this hypergroup (i.e. there exists a probability measure \(\mu\) such that \(P\{S_{n+1}| S_ n=x\}=\mu *\epsilon _ x)\) with \(E(\mu)<\infty\), then it is shown that \(S_ n/n\) converges almost surely to \(E_ *(\mu):=\int ^{\infty}_{0}m_ 1d\mu\) where \(m_ 1\) is the solution of the equation \[ m_ 1''+m_ 1'A'/A=\lim _{x\to \infty}A'(x)/A(x)=:2p,\quad m_ 1'(0)=m_ 1(0)=0. \] In the special case \(p=0\) this means \(S_ n/n\to 0\) a.s. It is also proved that in the case \(E(\mu)=\infty\), \(p>0\) almost surely \(S_ n/n\to \infty\).