A Markov modulated continuous-time capture-recapture population estimation model (Q812857)

A continuous time multi-area capture-recapture population model of animals is proposed. The size of the population in each area is not known and the movement of the population between areas was assumed to be modulated by some hidden Markov chain. A sample of the animals are captured, tagged and then released into their habitats. A proportion of the tagged animals are recaptured according to some Poisson processes whose rates are a function of the size of the population tagged, the size of the population and the movement of the animals between different areas. Based on observing the number of the tagged animals and the number of recaptured animals, estimates for the distribution of the population between the different areas were proposed. Finally, the EM algorithm was used to update the estimates of the parameters of the model.