Skip to main content
×
×
Home

On the properties of processes associated with a markov branching process

  • V. G. Gadag (a1) and R. P. Gupta (a2)
Abstract

Consider a time-homogeneous Markov branching process. We construct reduced processes, based on whether the length of line of descent of particles of this process are (a) greater than or (b) at most equal to, τ units of time, for some fixed τ ≧ 0. We show that in both cases the reduced processes retain the branching property, but the latter does not retain the time homogeneity. We investigate finite-time and asymptotic properties of the reduced processes. Based on a realization of the original process and a realization of a reduced process, observed continuously over a time interval [0, T] for T > 0, we propose estimators for the different parameters involved, including qτ , the probability that the original process becomes extinct before τ units of time, and f (j)(qτ ), the jth derivative of the offspring probability generating function f(s) at q τ when q τ is known. We study the properties of these estimators and derive their asymptotic distributions, under the assumption that the original process is supercritical.

Copyright
Corresponding author
Postal address: Department of Mathematics and Statistics, Memorial University of Newfoundland, St. John's, Newfoundland, Canada, A1C 5S7.
∗∗ Postal address: Department of Mathematics, Statistics, and Computing, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5.
Footnotes
Hide All

On study leave from University of Poona, Pune, India.

Footnotes
References
Hide All
Athreya, K. B. and Keiding, N. (1977) Estimation theory for continuous-time branching process. Sankhya A 39, 101123.
Athreya, K. B. and Ney, P. E. (1972) Branching Processes. Springer-Verlag, Berlin.
Gadag, V. G. and Rajarshi, M. B. (1989) On processes associated with a super-critical Markov branching process. Technical Report No. 1989–4, Statistics Division, Dalhousie University, Halifax, Canada.
Reynolds, J. F. (1972) A theorem on Markov branching process. J. Appl. Prob. 9, 667670.
Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Journal of Applied Probability
  • ISSN: 0021-9002
  • EISSN: 1475-6072
  • URL: /core/journals/journal-of-applied-probability
Please enter your name
Please enter a valid email address
Who would you like to send this to? *
×

Keywords:

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 7 *
Loading metrics...

Abstract views

Total abstract views: 48 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 12th June 2018. This data will be updated every 24 hours.