Birth-death recurrence classification (source code)

= Birth-death recurrence classification

Consider a <birth-death process> on $\{0,1,2,\ldots\}$ with constant birth rate $\lambda>0$, constant death rate $\delta>0$ at every positive state, and no deaths from zero. The reflecting boundary is essential: the result changes if zero is an <absorbing state>. <Detailed balance> would give
$$
\pi_{n+1}=\frac{\lambda}{\delta}\pi_n.
$$
When $\lambda<\delta$, these weights normalize to a <Geometric distribution>, so the <irreducible Markov chain>[irreducible] chain is <positive recurrent Markov chain>[positive recurrent]. When $\lambda=\delta$, the <jump chain>[embedded jump chain] is the symmetric reflected <random walk>: it returns to zero <almost surely>, but its constant equilibrium weights have infinite sum, so the chain is <null recurrent Markov chain>[null recurrent]. When $\lambda>\delta$, its upward-biased embedded <random walk> has <probability> $\delta/\lambda<1$ of hitting zero from state one, making the chain <transient Markov chain>[transient]. The <holding rate of a continuous-time Markov chain>[holding rates] are bounded, so <nonexplosion of a continuous-time Markov chain>[nonexplosion] creates no additional qualification here.