A simple branching process approach to the phase transition in \(G_{n,p}\) (Q1953327)

Summary: It is well known that the branching process approach to the study of the random graph \(G_{n,p}\) gives a very simple way of understanding the size of the giant component when it is fairly large (of order \(\Theta(n))\). Here we show that a variant of this approach works all the way down to the phase transition: we use branching process arguments to give a simple new derivation of the asymptotic size of the largest component whenever \((np-1)^3n\to\infty\).