Premium
On Population Size Estimators in the Poisson Mixture Model
Author(s) -
Mao Chang Xuan,
Yang Nan,
Zhong Jinhua
Publication year - 2013
Publication title -
biometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.298
H-Index - 130
eISSN - 1541-0420
pISSN - 0006-341X
DOI - 10.1111/biom.12044
Subject(s) - estimator , jackknife resampling , statistics , poisson distribution , mathematics , bootstrapping (finance) , nonparametric statistics , population , population size , confidence interval , extremum estimator , m estimator , econometrics , demography , sociology
Summary Estimating population sizes via capture–recapture experiments has enormous applications. The Poisson mixture model can be adopted for those applications with a single list in which individuals appear one or more times. We compare several nonparametric estimators, including the Chao estimator, the Zelterman estimator, two jackknife estimators and the bootstrap estimator. The target parameter of the Chao estimator is a lower bound of the population size. Those of the other four estimators are not lower bounds, and they may produce lower confidence limits for the population size with poor coverage probabilities. A simulation study is reported and two examples are investigated.