2013/11/15 by Kiranmoy Chatterjee, Chatterjee, Kiranmoy, Diganta Mukherjee +1
Computer Science · Mathematics · #62F15 #62P25 #Bayesian Methods and Mixture Models #Census and Population Estimation #FOS: Computer and information sciences #Methodology (stat.ME) #Statistical Methods and Bayesian Inference
paper · pdf · doi:10.48550/arxiv.1311.3812
openalex publication_date 2013/11/15 · openalex created_date 2022/10/05 · openalex updated_date 2026/07/28
For Dual-record system, in the context of human population, the popular\nChandrasekar-Deming model incorporates only the time variation effect on\ncapture probabilities. How-ever, in practice population may undergo behavioral\nchange after being captured first time. In this paper we focus on the\nDual-record system model (equivalent to capture- recapture model with two\nsampling occasions) with both the time as well as behavioral response\nvariation. The relevant model suffers from identifiability problem. Two\napproaches are proposed from which approximate Bayes estimates can be obtained\nusing very simple Gibbs sampling strategies. We explore the features of our two\nproposed methods and their usages depending on the availability (or\nnon-availability) of the information on the nature of behavioral response\neffect. Extensive simulation studies are carried out to evaluate their\nperformances and compare with few available approaches. Finally, a real data\napplication is provided to the model and the methods.\n