Sampling a branching tree (Q1263205): Difference between revisions

This paper is concerned with the sampling of a Galton-Watson branching tree, where family sizes of individuals selected at random from the population are recorded. Exact and limiting joint distributions of the sampled family sizes of two sampled individuals are derived; the probability of selecting different families is shown to tend rapidly to 1. Sampling from each of \(T+1\) consecutive generations is considered, and various inference techniques are presented, including parametric inference for the case of power series offspring distributions. Some properties of the maximum likelihood estimator (MLE) are discussed; for power series distributions, approximate MLEs are shown to be asymptotically normal and efficient.