Granville Tunnicliffe-Wilson is a Visiting Erskine Fellow at the University of Canterbury from 9th September to 8th December this year. The Erskine Fellowships are used to bring distinguished visitors from around the world to the University of Canterbury. Granville has the position of Reader in Statistics at the University of Lancaster and is author of the time series component of GenStat.

Darryl’s project is collaborative with Dr Jennifer Brown, University of Canterbury, James Nichols, Patuxent Wildlife Research Center, US Geological Survey and Andrew Royle, US Fish & Wildlife Service.

The proportion of sites (or more generally, area) occupied by a target species is of interest in many wildlife monitoring programmes. However, few species are so conspicuous that they will always be detected when present at a location. While detection of the species indicates the species is indeed present, non-detection does not equate to species absence.

A statistical model has recently been developed that enables estimation of the proportion of sites occupied by the species in such situations. This method requires a sampling scheme where N sites are surveyed at T (>1) discrete occasions, and the detection/nondetection of the species is recorded on each visit. The model provides a flexible framework for assessing such data, allowing relationships between occupancy and site characteristics (e.g. habitat type) to be investigated.

This research project is within the emerging interdisciplinary field of "astrostatistics", at the frontiers of physics and statistics. It is collaborative with Assoc Prof Nelson Christensen, Carleton College, Minnesota, and has strong links to Caltech, the University of California, Davis, Washington State University and the NSF-supported Laser Interferometer Gravitational Wave Observatory (LIGO) Scientific Collaboration. We will investigate novel statistical strategies for parameter estimation in complex astrophysical models. Making use of new simulation-based technology for posterior computation, known as Markov Chain Monte Carlo (MCMC), we will show that a Bayesian astrostatistical approach holds great potential and is no longer computationally prohibitive.

Two areas of astrophysics entering exciting phases of discovery and raising many statistical methodological issues, are:

· cosmic microwave background (CMB) cosmology: The NASA explorer mission MAP will measure temperature anisotropies in the CMB radiation, the remnant of radiation from the Big Bang, with unprecedented accuracy and provide data that can answer fundamental questions about the origin and fate of the universe, and

· gravitational radiation (GR) astronomy: A number of collaborations around the world are starting to operate laser interferometric GR antennae and may soon observe interesting astrophysical phenomena, e.g. coalescing binaries containing neutron stars or black holes. The detection of coalescing binary events will provide physicists with extremely useful cosmological information, such as an independent determination of the Hubble constant.

The extraction of cosmological and astrophysical parameters from the CMB and GR data requires a long and computer-intensive analysis chain to which Bayesian MCMC techniques will make a major contribution.

This project offers two PhD scholarships in statistics, and is looking for highly motivated graduate students in statistics (and/or maths/physics), preferably with a strong background in Bayesian inference and Monte Carlo methods, sound computing skills and a keen interest in applied-methodological, cross-disciplinary research. More details can be obtained from Dr Renate Meyer (meyer@stat.auckland.ac.nz, 09-373-7599 ext 5755).

As we all know, modern mathematics is a science of huge and complex structure. Although all mathematicians speak the same language and follow common rules, still mathematics, as a science, is very diverse. It resembles a great and extremely beautiful palace consisting of different wings with different structures, each with wonders of its own.

That is why mathematicians always appreciate research which aims to discover new connections between seemingly remote and unconnected wings of mathematical science. It seems that the Mathematical panel demonstrated that kind of attitude when they awarded the Marsden grant to the project "Local point processes in the neighbourhood of sets" (E. Khmaladze, Victoria University of Wellington). The project will explore very new and deep relationships between probability theory and the theory of random (point) processes on one side and differential and convex geometry on the other side. Although the main idea comes from New Zealand’s soil, two other researchers, W.Weil from Karlsruhe University, Germany, and J. Einmahl from Tilburg Catholic University, the Netherlands, will participate in the research as associate investigators.

In celebration of its 25th anniversary, the IASS established the Cochran- Hansen Prize to be awarded for the best paper on survey research methods submitted by a young statistician from a Developing or Transition Country. The next paper will be presented at the 54th Session of the International Statistical Institute, to be held in Berlin, Germany from August 13-20, 2003.

Participation in the competition for the Cochran-Hansen Prize is open to nationals of Developing or Transition Countries who are living in such countries and who were born in 1963 or later. Winners of an ISI Jan Tinbergen Award are not eligible for the competition.

Papers submitted must be unpublished original works. They may include materials from the participant’s university thesis. They should be in either English or French. The papers should be submitted to the IASS Secretariat at the address below, to arrive by 31 December 2002. Each submission should be accompanied by a cover letter that gives the participant’s year of birth, nationality, and country of residence.

The papers submitted will be examined by the Cochran-Hansen Prize Committee. The decision of the Committee is final.

The author of the winning paper will receive the Cochran-Hansen Prize in the form of books and journal subscriptions to the value of about 500 Euros and will be invited to present the paper at the Berlin Session of the ISI with all expenses paid (i.e., round trip airfare between place of residence and Berlin and a lump sum to cover living expenses).

For further information, please write to:
Madame Claude Olivier, IASS Secretariat
International Association of Survey Statisticians

CEFIL-INSEE,
3 rue de la Cité,
33500 Libourne, France
Tel: +33 5 57 55 56 17;
Fax: +33 5 57 55 56 20
E-mail: Claude.olivier@insee.fr