On the speed of coming down from infinity for subcritical branching processes with pairwise interactions

In this paper, we study the phenomenon of coming down from infinity for subcritical cooperative branching processes with pairwise interactions (BPI processes) under suitable conditions. BPI processes are continuous-time Markov chains that extend classical branching models by incorporating additional mechanisms accounting for both competitive and cooperative interactions between pairs of individuals. Our main focus is on characterising the speed at which BPI processes evolve when starting from a very large initial population in the subcritical regime. In addition, we investigate their second-order fluctuations. Furthermore, our results also apply to a class of exchangeable fragmentation-coalescent processes introduced by Berestycki (2004) and several other models from population genetics.