A0217
Title: An adaptive Monte Carlo method to estimate confidence interval for population sizes under mark-recapture-mark sampling
Authors: Ivair Silva - Federal University of Ouro Preto (Brazil)
Debanjan Bhattacharjee - Utah Valley University (United States) [presenting]
Yan Zhuang - Connecticut College (United States)
Abstract: The conventional mark-recapture strategy is modified to estimate the size ($N$) of a finite population. In this new procedure non-marked, resampled items are marked before they are released back into the population. A sequential adaptive stopping rule for fixed-length-interval-estimation of $N$ is proposed. A Monte Carlo solution is derived and compared with the accelerated sequential method. Estimating sizes of finite populations can become particularly relevant in knowing the total number of patients infected with a disease at a particular time in a geographical region. The new method is illustrated with a simulation that estimates the number of infected COVID-19 individuals in a near-closed population. In addition, we present a numeric application inspired by the problem of estimating the population size of endangered monkeys of the Atlantic Forest in Brazil.
