Research Priorities

• Marine ecology - how do we estimate animal movements using information contained in family trees?
• Population models - how do we efficiently calibrate Markovian models from discrete-sampled abundance data?

Researchers

• Chief Investigator: Phil Pollett
• Research Fellow: Ross McVinish
• PhD student: Robert Cope

Collaborating Researchers

• Dr Chantal Guihenneuc-Jouyaux, Université Paris Descartes
• Dr Janet Lanyon, School of Biological Sciences, University of Queensland
• Prof Kerrie Mengersen, Queensland University of Technology
• Dr Darfiana Nur, University of Newcastle
• Prof Judith Rousseau, Université Paris Dauphine
• Dr Jennifer Seddon, School of Veterinary Science, University of Queensland
• Prof Christian Weiss, Department of Mathematics and Statistics, Helmut Schmidt University, Hamburg

Selected Research Projects

• Recentered Importance Sampling

  Project leader and researcher: Ross McVinish (UQ)
  Researchers: Chantal Guihenneuc-Jouyaux (Université Paris Descartes), Kerrie Mengersen (Queensland University of Technology), Darfiana Nur (University of Newcastle), and Judith Rousseau (Université Paris Dauphine)

  Since its introduction in the early 1990s, importance sampling (IS) in concert with Markov chain Monte Carlo (MCMC) has found many applications. We have examined problems concerning its application to repeated evaluation of related posterior distributions, with a particular focus on Bayesian model validation. We have demonstrated that, in certain applications, the curse of dimensionality can be reduced by a simple modification of IS. In addition to providing new theoretical insight into the behavior of the IS approximation in a wide class of models, our result facilitates the implementation of computationally intensive Bayesian model checks. We illustrate the simplicity, computational savings, and potential inferential advantages of the proposed approach through two substantive case studies, notably computation of Bayesian p-values for linear regression models and simulation-based model checking.

  Research outputs

  • McVinish, R., Mengersen, K., Nur, D., Rousseau, J. and C. Guihenneuc-Jouyaux (2013) Recentered importance sampling with applications to Bayesian model validation. Journal of Computational and Graphical Statistics 22, 215-228.
• Calibration of Chain-Binomial Models with Density Dependent Thinning

  Project leader: Phil Pollett (UQ)
  Researcher: Christian Weiss (Helmut Schmidt University, Hamburg)

  We have developed an elaboration of the usual binomial AR(1) process on {0,1,...,N} that allows the thinning probabilities to depend on the current state n only through the "density" n/N, a natural assumption in many real contexts. We derive some basic properties of the model and explore approaches to parameter estimation. Some special cases are considered that allow for over- and underdispersion, as well as positive and negative autocorrelation. We derive a law of large numbers and a central limit theorem, which provide useful large-N approximations for various quantities of interest.

  Research outputs

  • ² Weiss, C.H. and P.K. Pollett (2014) Binomial autoregressive processes with density dependent thinning. Journal of Time Series Analysis 35, 115-132.

  ² To be reported in connection with ARC Discovery Grant DP110101929 "New Methods for Improving Active Adaptive Management in Biological Systems".

• Genetic Marker Based Pedigree Reconstruction

  Project leader: Phil Pollett (UQ)
  Researchers: Robert Cope (UQ), Janet Lanyon (UQ), and Jennifer Seddon (UQ)

  For wildlife populations it is often difficult to determine biological parameters that indicate breeding patterns and population mixing, but knowledge of these parameters is essential for effective management. A pedigree encodes the relationship between individuals, and can provide insight into the dynamics of a population over its recent history. We have developed a method for the reconstruction of pedigrees for wild populations of animals that live long enough to breed multiple times over their lifetime and that have complex or unknown generational structures. Reconstruction was based on microsatellite genotype data along with ancillary biological information: sex, and observed body size class as an indicator of relative age of individuals within the population. Using body size-class data to infer relative age has not been considered previously in wildlife genealogy, and provides a marked improvement in accuracy of pedigree reconstruction. Body size class data is particularly useful for wild populations because it is much easier to collect non-invasively than absolute age data. This new pedigree reconstruction system, PR-genie, performs reconstruction using maximum likelihood with optimization driven by the Cross-Entropy method. We demonstrated pedigree reconstruction performance on simulated populations (comparing reconstructed pedigrees to known true pedigrees) over a wide range of population parameters and under assortative and intergenerational mating schema. Reconstruction accuracy increased with the presence of size-class data and as the amount and quality of genetic data increased. We arrived at recommendations as to the amount and quality of data necessary to provide insight into detailed familial relationships in a wildlife population using this pedigree reconstruction technique.

  Research outputs

  • ³ Cope, R.C., Lanyon, J.M., Seddon, J.R. and P.K. Pollett (2014) Development and Testing of a Genetic Marker Based Pedigree Reconstruction System `PR-genie' Incorporating Size-Class Data. Molecular Ecology Resources 14, 857-870.

  ³ To be reported in connection with ARC Linkage Grant LP0882316 "Animal Movement Between Populations Deduced From Family Trees - A Test Case on Dugongs in Southern Queensland".