Resampling procedures for assessing the fit of capture-recapture models

Lines connect the coordinator with its partners.

Project objective

Research objectives and content The project constitutes a key addition to recent methodological advances, which have enabled capture-recapture modelling to be undertaken using an analysis-of-variance approach. This approach involves first fitting an all-embracing 'umbrella model', which is then successively refined to provide a parsimonious description of the data. At present, however, this major methodological advance can founder due to the lack of reliable goodness-of-fit tests of the initial model. The project aims to develop resampling procedures to fill this gap. The first objective will be to provide bootstrap or Monte-Carlo based goodness-of-fit tests for the Cormack-Jolly-Seber family. These should be a major improvement on the existing asymptotic procedures. The second objective will be to extend these methods to encompass recently-developed multi-site models which are already being seen to be extremely useful in practice. Training content (objective, benefit and expected impact) Whilst working within the group at Montpellier I will be in an ideal position to extend my solid background in traditional capture-recapture methodology to cover both their analysis-of-variance approach and an increased range of applications of the material. The project will extend my experience of the resampling techniques in which my current University (St Andrews) has a leading position. Links with industry / industrial relevance (22) NA

Original text from CORDIS.

Participants

• CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE · PARISCoordinatorFrance

Links

• View on CORDIS