Honorary Life Members

Honorary Life Membership of the NZSA was granted to Steve Kuzmicich and John Darwin, who have both been Government Statisticians in the course of their illustrious careers. The text below is adapted from citations they received from the Association.

Steve Kuzmicich served as Secretary-Treasurer of NZSA from 1959-1960 and as President from 1963-1965 and contributed very widely to the Association since joining in 1956. Next year he celebrates 50 years as a member of the NZSA. He has contributed to statistics in New Zealand on the wider sphere, particularly as Government Statistician from 1984–1991, and has helped raise statistical awareness within the community. He has earned his role as the Elder Statesman of Statistics in NZ.

John Darwin (right) served as Secretary-Treasurer of the NZSA from 1956-8 and as President from 1961-3. He was leader of the Statistics Section of the Applied Mathematics Division of DSIR for many years. He was appointed in 1978 as Deputy Government Statistician and then in 1980 as Government Statistician, head of the Department of Statistics, until his retirement in 1984, the culmination a very successful career. A fascinating biographical note on John’s career, written by Robert Davies in 1984, is available at http://ifs.massey.ac.nz/mathnews/
centrefolds/31/Aug1984.shtml.

Marsden Fund Awards - Beatrix Jones, David Bryant

Congratulations to Beatrix Jones (Massey University, Albany Campus) and David Bryant (University of Auckland), who received Marsden Fund Awards in this year's round.

Beatrix's project is "Design of parentage analysis experiments: a case study for understanding uncertainty in models with latent (unobserved) variables", and is a two year Fast Start award. Beatrix (right) welcomes inquiries from prospective domestic postgraduates who would like to do a funded honours project in this area.  Parentage analysis refers to a group of techniques ecologists use to understand mating and dispersal patterns of wild organisms.  In a parentage analysis, genetic data are collected from a group off offspring, and adults that may be their parents.  The genetic data is then used to match up offspring with their parents; questions like "do older adults produce more offspring" can then be answered. Unfortunately, it's difficult to collect enough genetic data to do this parent-offspring matching exactly.  That's where statisticians come in. When not all offspring can be confidently assigned to parents, the parent assignments can be treated as latent variables.  Demographic parameters can then be estimated by sampling from the joint posterior of demographic parameters and parent assignments.   When this method is used, the relationship between the amount of data collected and the precision of the resulting parameter estimates is complex.  Beatrix's project will develop design methods so ecologists can know before they begin data collection how much data they will need to achieve a desired precision.  She hopes to extend these methods to other contexts as well.

David's project is on " The statistics of phylogenetic networks". Split networks, and phylogenetic networks, are data representation tools that have arisen out of research into phylogenetic analysis (reconstruction of evolutionary history), but have been applied to fields as diverse as comparative linguistics (see Figure) and virology. In effect, they can be regarded as compact representations of large collections of trees. The central aim of this project will be to place these representation tools into an appropriate statistical framework, with the development of the appropriate network models, validation and sampling techniques. These will be applied to the analysis of complex evolutionary events: particularly the study of viral transmission histories and evolutionary analysis of organisms with unclear species boundaries. Project members include Alexei Drummond (Auckland), Noah Rosenberg (Michigan), Bernd Sturmfels (Berkeley) and Paul Tupper (McGill). Software is available at www.splits.org. David is  currently searching for two Ph.D. students.

NZIMA Programme - Modelling Invasive Weed Species

Congratulations to Jennifer Brown, Alex James and David Wall (U Canterbury) who are Directors of an NZIMA Programme on "Modelling Invasive Species and Weed Impact", proposed to start late 2006. The aim of the 3-year programme is to bring mathematicians and statisticians together with biologists to stimulate applied research that will benefit weed control and management.

The programme will begin with a 5 day workshop in Hanmer in December 2006. Up to 6 international mathematicians and statisticians, along with about 35 New Zealanders, will be invited. The workshop's format will be introductory sessions by New Zealand weed managers outlining the current issues and problems in weed management in NZ, followed by sessions from the international invitees on the latest developments in relevant mathematical and statistical tools. Each day, in the follow-up sessions, the workshop attendees will identify the gap between the knowledge that can be gained from the the current mathematical models and what is needed by NZ weed managers. The NZIMA programme goal is to bridge that gap by stimulating relevant research amongst NZ mathematicians and statisticians.

Applications are being called for students to undertake postgraduate and postdoctoral studies in this area. The NZIMA programme will also include follow-up workshops and regular newsletter style updates on the research.

For more information or for any queries contact Jennifer Brown at the Biomathematics Research Centre at University of Canterbury.

NZSA Campbell Award Criteria

Jennifer Brown