Pascal Lecture 2017
Professor Peter Diggle (Lancaster)
"Statistical Analysis of Spatiotemporal Point Process Data"
Abstract: A spatiotemporal point process, P, is a stochastic model for generating a countable set of points (x(i), t(i)) ∈ IR2 × IR+, where each x(i) denotes the location, and t(i) the time, of an event of interest. A typical data-set is a partial realisation of P restricted to a specified spatial region A and time-interval [0,T], possibly supplemented by covariate information on location, time or the events themselves. In this talk, I will first give examples of different interpretations of this scenario according to whether only one or both of the sets of locations and times are stochastically generated. I will then discuss in more detail methods for analysing spatiotemporal point process data based on two very different modelling approaches, log-Gaussian Cox process models; and conditional intensity models, and describe applications of each in the context of human and veterinary epidemiology.
Peter Diggle is a Fellow of the Royal Statistical Society and Distinguished University Professor in the Lancaster Medical School, and holds a part-time post at the University of Liverpool, Department of Epidemiology and Population Health. He also has adjunct appointments at the Johns Hopkins University School of Public Health, Columbia University International Research Institute for Climate and Society, and Yale University School of Public Health. He is a trustee for the Biometrika Trust, a member of the Advisory Board for the journal Biostatistics, chair of the Medical Research Council’s Strategic Skills Fellowship Panel and President-Elect of the Royal Statistical Society.
His research concerns the development and application of statistical methods relevant to the biomedical and health sciences.