Back to search

University of Southampton

Capture-recapture estimation and modelling for one-inflated count data

Abstract

dc:description.abstract

Capture-recapture methods are used to estimate the unknown size of a target population whose size cannot be reasonably enumerated. This thesis proposes the estimators and the models specifically designed to estimate the size of a population for one-inflated capture-recapture count data allowing for heterogeneity. These estimators can assist with overestimation problems occurring from one-inflation that can be seen in several areas of researches. The estimators are developed under three approaches.<br/>The first approach is based on a modification by truncating singletons and applying the conventional Turing and maximum likelihood estimation approach to the one-truncated geometric data for estimating the parameter <i>p</i><sub>0</sub>. These <i>p</i><sub>0</sub> are applied to the Horvitz-Thompson approach for the modified Turing estimator (T_OT) and the modified maximum likelihood estimator (MLE_OT).<br/>The second approach is the model-based approach. It focuses on developing a statistical model that describes the mechanism to generate the extra of count ones. The new estimator MLE_ZTOI is developed from a maximum likelihood approach by using the nested EM algorithm based upon the zero-truncated one-inflated geometric distribution<br/>The last approach focuses on modifying a classical Chao’s estimator to involve the frequency of counts of twos and threes instead of the frequency of counts of ones and twos. The modified Chao estimator (MC) is asymptotic unbiased estimator for a power series distribution with and without one-inflation and provides a lower bound estimator under a mixture of power series distributions with and without one-inflation. The three bias-correction versions of the modified Chao estimator have been developed to reduce the bias when the sample size is small. A variance approximation of MC and MC3 are also constructed by using a conditioning technique.<br/>All of the proposed estimators are assessed through simulation studies. The real data sets are provided for understanding the methodologies.

Degree

thesis:*
Name dc:type.qualificationname
Ph.D.
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Southampton
Year dc:date.issued
2018

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Kaskasamkul, Panicha
Advisor dc:contributor.advisor
  • Bohning, Dankmar
ORCID person:identifier.orcid

Rights

Language dc:language
English, eng

Chain of custody

source
Harvested from
University of Southampton
Base URL
eprints.soton.ac.uk/cgi/oai2
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Kaskasamkul, Panicha. Capture-recapture estimation and modelling for one-inflated count data. doctoral thesis, University of Southampton, 2018.