Statistics Across Campuses

Dealing with detection error in site occupancy and abundance surveys: what can we do with a single survey? Date: 04 December 2020, Friday Time: 10am AEDT  Speaker: Prof Subbash Lele  (University of Alberta) Abstract: Site occupancy probabilities and abundance of target species are commonly used in various ecological studies. Detection error introduces bias in the estimators of site occupancy. Existing methods for estimating occupancy probability in the presence of detection error use replicate surveys. These methods assume either population closure, i.e. the site occupancy status remains constant across surveys or a structured dependence across surveys. The practical difficulties with replicate surveys are well known. Statistically, the closure assumption is seldom satisfied and the dependence structure can be difficult to model. The cost of replicate surveys can be prohibitive and they may be logistically prohibitive.  Given these practical difficulties, we ask the question: what can

Computing Bayes: Bayesian Computation from 1763 to the 21st Century Date: 19 November 2020, Thursday Time: 10.30am AEDT Contact the organizer: Andriy Olenko a.olenko@latrobe.edu.au Speaker: Prof Gael M. Martin (Monash University) Abstract: The Bayesian statistical paradigm uses the language of probability to express uncertainty about the phenomena that generate observed data. Probability distributions thus characterize Bayesian inference, with the rules of probability used to transform prior probability distributions for all unknowns - models, parameters, latent variables - into posterior distributions, subsequent to the observation of data. Conducting Bayesian inference requires the evaluation of integrals in which these probability distributions appear. Bayesian computation is all about evaluating such integrals in the typical case where no analytical solution exists. This paper takes the reader on a chronological tour of Bayesian computation over the past two and a half centuries. B

Change-set analysis for spatial clustering in environmental health Date: 20 November 2020, Friday Time: 10am AEDT  Speaker: Prof Jun Zhu  (University of Wisconsin-Madison) Abstract: Mapping of disease incidence is of importance to environmental health. In this talk, we consider identification of clusters of spatial units with elevated disease rates and develop a new approach that estimates the relative disease risk in association with potential environmental risk factors and simultaneously identifies clusters corresponding to elevated risks. A heterogeneity measure is proposed to enable the comparison of a candidate cluster and its complement under a pair of complementary models. A quasi-likelihood procedure is developed for estimating the model parameters and identifying the clusters. An advantage of our approach over traditional spatial clustering methods is the identification of clusters that can have arbitrary shapes due to abrupt or non-contiguous changes while accounting for risk