The residual lifetime process of a renewal process is the time remaining until the next renewal:
The time since the last renewal is , and is the length of the renewal interval containing a fixed time. Between renewal epochs, decreases at speed one. On reaching zero it resets to a freshly drawn interarrival time. For strictly positive iid interarrivals, this deterministic motion and independent resetting make a Markov process, even when the count alone is not Markov.
Its initial and equilibrium laws can differ. An ordinary process started at a renewal has . A stationary observation instead has the equilibrium residual-life distribution, whose survival function is the integrated interarrival tail divided by its mean. For deterministic interarrivals of length , the initial residual is , but the equilibrium residual is uniform on . For exponential interarrivals the two laws agree, reflecting the memoryless property.
For non-arithmetic inter-renewal times with finite positive mean , the excess converges in distribution to with
Suppose an independent superposition of a positive-rate Poisson process and a renewal process with strictly positive nonarithmetic finite-mean interarrivals is itself a renewal process. Its first interarrival must be exponential; in fact the other process's interarrival law must also be exponential.
Let be the Poisson rate, the other interarrival survival function and its finite mean. The first superposed interarrival has survival and mean . Adding the two long-run counting rates gives . Computing the limiting excess in two ways, using the renewal excess limit theorem for the superposition and independence of the two excesses, gives
Set . Differentiating the displayed identity almost everywhere yields . Thus with , forcing . It follows that . The renewal assumption on the superposition is essential; independent superpositions of general renewal processes need not be renewal processes.
For a positive integer-valued lifetime , one discrete-time residual lifetime Markov chain uses states . From state it moves to . From zero it draws an independent new lifetime and moves to :
The subtraction of one specifies the sampling convention: the next step has already consumed one unit of the new lifetime. With this convention, successive visits to zero are separated by exactly steps, not .
Its Stationary distribution is . The tail-sum identity normalizes these masses, and the balance equations verify them, with the obvious top-state equation. If deterministically, the chain cycles and its stationary law is uniform. It is periodic, so a Stationary distribution does not imply convergence from every initial state.
The equilibrium residual-life distribution describes the remaining wait to a renewal when observing a stationary renewal process at an ordinary time. For positive interarrivals with finite mean , it has survival function and density
The density integrates to one by the Tail-sum formula for expectation. It exists even when the original lifetime law has atoms. A deterministic lifetime gives a uniform residual on ; an exponential lifetime gives the same exponential distribution as the original lifetime.
For the discrete chain convention in residual lifetime Markov chain, the invariant probabilities are instead for . A deterministic integer lifetime gives a uniform law on . These formulas describe observation at equilibrium; they do not make the first lifetime of an ordinary renewal process already have that residual law.
For inter-renewal times with an Erlang distribution of integer shape and rate , the mean is and survival function is . The equilibrium first-delay density is survival divided by the mean,
This is the equal mixture of Erlang distributions of shapes . It can be constructed by observing a Poisson clock with uniformly randomized phase modulo , whose every th event is a renewal.

Articles by others on the same topic (0)

There are currently no matching articles.