You are cordially invited to participate in a workshop addressing the Modelling of individual histories with state uncertainty, 20-24 March 2017
The workshop will be held in Montpellier (France) on the CNRS campus and will be taught by Roger Pradel, Rémi Choquet, Olivier Gimenez, Lorelei Guery, Meritxell Genovart.
Application deadline is January 19.
The estimation of population parameters (survival, recruitment, dispersal, and population growth rate) is critical to many areas of fundamental and applied biology. The major source of data for such estimation comes from observations of marked animals. This workshop will deal with recent advances in the analysis of such data.
The content is aimed at providing the participants with a solid background in the philosophy, theory and practices for the analysis of data from marked animals, with a specific focus on multievent data and models. Multievent capture-recapture models are a natural generalization of the multisite recapture models. Similarly, individuals are sampled on discrete occasions, at which they may be captured or not. However, contrary to the multisite case, uncertainty in the assessment of state such as breeder, diseased or highly catchable can be incorporated into the analysis. Conceptually, it is not states that are observed but rather something dubbed an "event" (a particular breeding behaviour, a positive blood test or just encounter), which reflects to some extent the underlying state. The presence of imperfect observations and the lack of information on individual quality presently make multievent models an invaluable tool in virtually any area of population biology. Current applications include the study of dispersal, epidemiology, individual heterogeneity, mixture of information... New developments allow the treatment of individual heterogeneity through random effects as well as through mixture models, and the inclusion of individual as well as environmental covariates. For examples of topics that can be addressed in the multievent framework, see the non-exhaustive bibliography here.
Emphasis will be placed on stringent procedures for building and selecting the most appropriate models for the data set at hand, in order to be able to draw reliable biological conclusions.
Format of the workshop will be a combination of lectures and computer lab exercises with the programs E-SURGE and U-CARE. U-CARE incorporates goodness-of-fit tests for multistate models, and E-SURGE makes the building of complex models easy based on a language describing the structure of models in a compact and user-friendly way. See here for more details about the installation of these pieces of software. All lectures and course material will be in English.
It is our hope that participants will bring their own capture-recapture data to work on them during the workshop. To take full advantage of the workshop, it is expected that the participants will have some basic practice of the analysis of CR data.
Participants will have to bring their own laptop that is capable of running Windows-based software.