Covariates have been used to model parameters in capture-recapture methods in ecology since the paper by North and Morgan (1979). In this talk we consider a variety of issues that remain to be answered, including how to select covariates, how to deal with spatial-variation information and how to include time-variation in recovery probabilities for conditional analysis of ring-recovery data.
The talk describes research with Dan Brown, Diana Cole, and Ian Jolliffe of the National Centre for Statistical Ecology, and David Thomson of the University of Hong Kong. Illustrative applications are provided by white stork, Ciconia ciconia and blackbird, Turdus merula. The methods described include spatial splines, parameter redundancy and the lasso, all of which will be presented in outline only.
Cole, D.J. and Morgan, B.J.T. (2010) Parameter redundancy with covariates, Biometrika, 97, 1002-1005.
Cole, D.J., Morgan, B.J.T. and Titterington, D.M. (2010) The parametric structure of models, Mathematical Biosciences, 228, 16-30.
North, P.M. and Morgan, B.J.T. (1979) Modelling heron survival using weather data, Biometrics, 35, 667-682.
Keywords: Ecology; Capture-recapture; parameter redundancy; covariates
Biography: Byron Morgan is Head of the Statistics Division at the University of Kent, and co-Director of the National Centre for Statistical Ecology in the UK. He is Vice President of the British and Irish Region of the International Biometric Society. His research interests include stochastic modelling in Statistical Ecology and in Molecular Biology. He has published books on Simulation, Analysis of Quantal Response Data, and Applied Stochastic Modelling, and he has co-authored a book on Bayesian Analysis for Population Ecology. In the past, he has edited Applied Statistics, Biometrics Shorter Commuinications and JABES.