Parameterised branching processes: a functional version of Kesten \& Stigum theorem (Q2169076)

Let \(I \subseteq [0,\infty)\) be an interval and \((X(\lambda))_{\lambda \in I}\) an almost surely non-decreasing, integer-valued stochastic process taking values in the Skorokhod space of càdlàg functions on \(I\) taking values in \([0,\infty)\). Assume that \(\mathbb{E}[X(\lambda)]=\lambda\) for every \(\lambda \in I\). Let \((X^{(k,n)})_{k, n \in \mathbb{N}}\) be a family of independent copies of \(X\). Now define \(Z_0(\lambda)=1\) for all \(\lambda \in I\) and, recursively, \[ \qquad \qquad \quad Z_{n+1}(\lambda) = \sum_{k=1}^{Z_{n}(\lambda)} X^{(k,n)}(\lambda), \quad n \in \mathbb{N}_0 = \{0,1,2,3,\dots\}. \] The authors call \((Z_n(\lambda))_{(n,\lambda) \in \mathbb{N}_0 \times I}\) a Galton-Watson random field. Here, \(Z_n(\lambda)\) is seen as the size of generation \(n\) at time \(\lambda\). The classical martingale in the Galton process here is \[ W_n(\lambda) = \frac{Z_n(\lambda)}{\mathbb{E}[X(\lambda)]^n}, \quad n \in \mathbb{N}_0. \] For any fixed positive \(\lambda \in I\), \((W_n(\lambda))_{n \in \mathbb{N}_0}\) is a non-negative martingale and, therefore, converges almost surely to some non-negative random variable \(W(\lambda)\) with \(\mathbb{E}[W(\lambda)] \leq 1\). In the paper under review, the authors address the question of whether there is stronger convergence than pointwise almost certain convergence. The attention is restricted to the supercritical case, i.e., \(I \subseteq (1,\infty)\), as the martingale limit \(W(\lambda)\) vanishes almost surely if \(\lambda \leq 1\). The authors show, under an assumption that gives control over the mixed factorial moments of the increments of \(X\), that \((W_n(\lambda))_{\lambda \in I}\) converges in probability in the Skorokhod space to a limiting process \(W=(W(\lambda))_{\lambda \in I}\). The limiting process \(W\) satisfies the functional analogue of the classical fixed-point equation satisfied by the limit of the martingale in the Galton-Watson process.