Full Likelihood Inference for Abundance from Capture-Recapture Data

Explore a comprehensive lecture on full likelihood inference for abundance estimation using capture-recapture data. Delve into the semiparametric model proposed by Huggins (1989) and Alho (1990) that accounts for heterogeneity in capture probabilities. Examine the limitations of the conditional likelihood method and discover a new full likelihood approach based on Huggins and Alho's model. Learn about the asymptotic properties of the empirical likelihood ratio and the semiparametric efficiency of the maximum empirical likelihood estimator. Investigate the proposed expectation-maximization algorithm for numerical calculations and compare the performance of empirical-likelihood-based and conditional-likelihood-based methods through simulation studies. Gain insights into improved confidence interval coverage and reduced mean square error in abundance estimation for closed populations.