Design problems for the pure birth process
- 1 June 1983
- journal article
- Published by Cambridge University Press (CUP) in Advances in Applied Probability
- Vol. 15 (2) , 255-273
- https://doi.org/10.2307/1426436
Abstract
LetY(t) be a pure birth process. If a maximum likelihood estimator of the birth intensity is desired and the numbernof observational points and the last observation T are given in advance, it is shown that equidistant sampling is not an optimal procedure. Properties of ‘optimal' designs and the corresponding maximum likelihood estimators are investigated and compared with equidistant and continuous sampling.Keywords
This publication has 8 references indexed in Scilit:
- FIXED VERSUS RANDOM SAMPLING OF CERTAIN CONTINUOUS PARAMETER PROCESSESAustralian Journal of Statistics, 1980
- Nonparametric Inference for a Family of Counting ProcessesThe Annals of Statistics, 1978
- Some aspects of the theory of estimating equationsJournal of Statistical Planning and Inference, 1978
- Maximum Likelihood Estimation in the Birth-and-Death ProcessThe Annals of Statistics, 1975
- Estimation in the birth processBiometrika, 1974
- An Optimum Property of Regular Maximum Likelihood EstimationThe Annals of Mathematical Statistics, 1960
- Efficiency Problems in Polynomial EstimationThe Annals of Mathematical Statistics, 1958
- Stochastic Processes and Population GrowthJournal of the Royal Statistical Society Series B: Statistical Methodology, 1949