Approximating persistence in a general class of population processes.

Journal article

We provide a general framework for estimating persistence in populations which may be affected by catastrophic events, and which are either unbounded or have very large ceilings. We model the population using a birth-death process modified to allow for downward jumps of arbitrary size. For such processes, it is typically necessary to truncate the process in order to make the evaluation of expected extinction times (and higher-order moments) computationally feasible. Hence, we give particular attention to the selection of a cut-off point at which to truncate the process, and we present a simple method for obtaining quantitative indicators of the suitability of a chosen cut-off.

Authors: Cairns, B; Pollett, P

• Publication status: Published
• Journal: Theoretical population biology
• Volume: 68
• Issue: 2
• Pages: 77-90
• Publication date: 2005-08-01
• DOI: 10.1016/j.tpb.2005.02.002
• EISSN: 1096-0325
• ISSN: 0040-5809
• Language: English
• Keywords: Insurance, Major Medical; Models, Statistical; Disasters; Birth Rate; Markov Chains; Mortality; Models, Theoretical; Population Dynamics