Applications of hidden hybrid Markov/semi-Markov models: from stopover duration to breeding success dynamics

摘要: Usually in capture–recapture, a model parameter is time or since first cap-ture dependent. However, the case where probability of staying one state depends on spent that particular not rare. Hidden Markov models are appropriate to manage these situations. A more convenient approach would be consider incorporate semi-Markovian states which explicitly define waiting distribution and have been used pre-vious biologic studies as framework for modeling given physiological state. Here, we propose hidden Markovian combine several nonhomogeneous with (i) well adapted imperfect variable detection (ii) allow us time, capture, effects. Implementation details depending number discussed. From user's perspective, present enhances toolbox analyzing capture–recapture data. We then show potential this by means two ecological examples: stopover duration breeding success dynamics.