Absorbing Markov and branching processes with instantaneous resurrection
A Markov branching process with instantaneous immigration from the zero state can be constructed so as to be honest and have the non-negative integers as state-space, but the construction requires the branching part to be explosive. We show that a realistic model can be constructed without this restriction if the state-space is restricted to the natural numbers. Moreover this construction is the weak limit, in the sense of finite dimensional laws, of the Yamazato model as the zero state holding-time parameter tends to infinity. This idea of immediate resurrection from an absorbing subset is extended to any minimal discrete-state Markov process, and even to a larger class. Our emphasis is on existence and uniqueness of the transition functions of the resurrected process, and classification of its states.
Year of publication: |
1993
|
---|---|
Authors: | Pakes, Anthony G. |
Published in: |
Stochastic Processes and their Applications. - Elsevier, ISSN 0304-4149. - Vol. 48.1993, 1, p. 85-106
|
Publisher: |
Elsevier |
Keywords: | branching and Markov processes transition functions and generators resurrection recurrence classification |
Saved in:
Saved in favorites
Similar items by person
-
Limit laws for UGROW random graphs
Pakes, Anthony G., (2013)
-
Characterization of discrete laws via mixed sums and Markov branching processes
Pakes, Anthony G., (1995)
-
Pakes, Anthony G., (1979)
- More ...