Premium
LOWER BOUNDS TO THE POPULATION SIZE WHEN CAPTURE PROBABILITIES VARY OVER INDIVIDUALS
Author(s) -
Xuan Mao Chang
Publication year - 2008
Publication title -
australian and new zealand journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 1369-1473
DOI - 10.1111/j.1467-842x.2008.00503.x
Subject(s) - mathematics , estimator , statistics , upper and lower bounds , range (aeronautics) , population size , population , confidence interval , maximum likelihood , construct (python library) , sequence (biology) , coverage probability , econometrics , demography , computer science , biology , mathematical analysis , materials science , sociology , composite material , genetics , programming language
Summary The problem of estimating population sizes has a wide range of applications. Although the size is non‐identifiable when a population is heterogeneous, it is often useful to estimate the lower bounds and to construct lower confidence limits. A sequence of lower bounds, including the well‐known Chao lower bound, is proposed. The bounds have closed‐form expressions and are estimated by the method of moments or by maximum likelihood. Real examples from epidemiology, wildlife management and ecology are investigated. Simulation studies are used to assess the proposed estimators.