General Harris regularity criterion for non-linear Markov branching processes (Q2489811)

Let \(a>0\) and \(\{p_j\); \(j\geq 0\}\) be a probability distribution defined on \(Z_+\) and such that \(p_0\times\sum_{j=2}^{\infty}p_j>0\). A matrix \(Q=\{q_{ij}\); \(i,j\in Z_+\}\) is called a nonlinear branching \(q\)-matrix with parameter \(\theta >0\) if \(q_{ij}=ai^{\theta}p_{j-i+1}\) for \(j\geq i-1,j\neq i,\) \(q_{ii}=-ai^{\theta}(1-p_1),\) and \(q_{ij}=0\) for \(j<i-1\). A nonlinear Markov branching process is a \(Z_+\)-valued continuous time Markov chain whose transition function \(P(t)=\{p_{ij}(t)\); \(i,j\in Z_+\}\) satisfies the Kolmogorov forward equations \(P^{\prime}(t)=P(t)Q\). The authors give a criterion for the nonlinear Markov branching process to be regular.