Work

Embeddability of Discrete Time Simple Branching Processes into Continuous Time Branching Processes

Year: 1968
Type: article
Abstract: A simple discrete time Markoff branching process {Xn, «^0} with state space the nonnegative integers is induced by a probability generating function (abbreviated hereafter as p.g.f.) f(x) = J.k°=0 akxk where ak^0, k=0, 1,2, .. .,/(l)= 1.The transition probability matrix Pii = Pr{Xn + x =j\Xn = i} is determined from the generating function by the relations 2 Pi,x'= [f(x)]K i = 0,1,2,.... 1 = 0 The p.g.f.f(x) is interpreted as the number of progeny produced by each parent per generation.All offspring act independently with the same fixed lifetime and the same distribution of progeny.It is a familiar fact that the « step transition probability matrix pp = Pr{Xn=j\X0 = i} possesses the generating function 2 P&* = [fn(x)Y i = o where fn(x)=fn-x(f(x)) is the «th functional iterate of/(x).The branching process is called critical, supercritical or subcritical according as f'(l) = m=l, m>l or m < 1.For our purposes a one-dimensional continuous time Markov branching process {Zt, t ^ 0} is a temporally homogeneous continuous time Markov chain with the nonnegative integers as state space, characterized by the following properties: (i) the p.g.f.,sft(x) = Jik=0 xk Pr {Zt=k | Z0 = 1} continuous in /^0 and |.x| S 1 are related functionally in the manner ft+s(x)=ft(fs(x)), t,s^0,\x\Sl,
Language: English
Open Access status: bronze
OpenAlex ID: W2129490488
FWCI: 1.24
Cites: 11
Cited by: 75
Related to: 10