{"id":"arxiv/2607.12564","url":"https://arxiv.org/abs/2607.12564","tags":["math.PR"],"score":0,"title":"Power-law and log-periodic degree tails for a family of probability generating function equations arising in evolving networks","domain":"math.PR","source":"arxiv","authors":["Qunqiang Feng","Xiao-Ming Fu","Tianyang Sun"],"reviews":0,"subject":"math.PR","summary":"For a fixed integer $j\\ge1$ and $0<p<1$, we study the probability generating function (pgf) equation \\[ (1+2p)\\,g(x)=2p\\,x^{j}+g\\bigl(x-px+px^{2}\\bigr),\\qquad 0\\le x\\le1 , \\] which governs the limiting degree distribution $\\{p_k\\}$ of a family of evolving network models. The cases $j=1$ and $j=2$ are the treelike fast-growth model of Feng and Hu and the homogeneous evolving network of Feng, Li and Hu. We prove that for every $j$ the equation has a unique pgf solution, of mean $2j$, and we determine its coefficient tail exactly: \\[ p_k=k^{-1-ρ}\\,Ψ_j(\\log_λk)+o\\bigl(k^{-1-ρ}\\bigr), \\] where…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-14","reactions":{}}