The central limit theorem for the supercritical branching random walk, and related results (Q916215): Difference between revisions

The standard branching random walk is a Galton-Watson process with one- dimensional spatial structure whereby the offspring of a parent take up positions relative to the parent according to a given point process, and different families behave independently. If \(Z^{(n)}\) denotes the n th generation point process (starting with a single ancestor at the origin) and \(N^{(n)}=Z^{(n)}({\mathbb{R}})\) is the size of the n th generation, then on the event \(\{N^{(n)}>0\}\), \(Z^{(n)}/N^{(n)}\) is a random probability measure on \({\mathbb{R}}\). The fact that the intensity measure of \(Z^{(n)}\) is an n-fold convolution suggests that this random probability measure behaves like that of the sum of n independent, identically distributed random variables, notably in respect of central limit or, more generally, stable properties. This paper generalizes and strengthens earlier partial results in this area, such as those of \textit{N. Kaplan} and \textit{S. Asmussen} [Stochastic Processes Appl. 4, 15-31 (1976; Zbl 0322.60065)] and \textit{C. F. Klebaner} [Adv. Appl. Probab. 14, 359-367 (1982; Zbl 0482.60078)].