Newsletter 66 Index

Australian &
New Zealand Journal of Statistics

Newsletter Archive

Join the NZSA

Feedback to Editor

New Zealand Statistical Association Newsletter 66

September 2007

Local News

AgResearch
We recently farewelled our leader Peter Johnstone, who retired from being section manager in June. Peter and his wife Anne have since been travelling overseas for a couple of months. Upon his return, Peter will work as a biometrician for two days a week at AgResearch Invermay. In Peter’s place, we welcome Anette Becher, who was previously the senior person in the bioinformatics part of our group. At this time, a review of our group is being carried out by Bryan Manly to help Anette crystal-ball-gaze into our future. One thing that is evident without the crystal ball, however, is that we are short of biometricians in our section, and are actively seeking out new staff.

Our annual section meeting and “team bonding” get-together (see photo - click to enlarge) was held in April at Ruakura, for two days following the Baddeley workshop at Waikato University. Almost all of us gave talks about our work. As light relief, we also tried to identify wines and fruit juices, some with notable success and most of us with notable failure!

Congratulations to Ken Dodds for being the 2007 recipient of the prestigious McMeekan Memorial Award, presented by the New Zealand Society of Animal Production for his contribution to animal production in New Zealand. These contributions include the development of the fractional-parentage DNA pedigree system and contributions to gene discovery programmes. Great work, Ken!

Fred Potter was co-author on a prize-winning paper entitled “Drenching adult ewes: implications of anthelmintic treatments pre- and post-lambing on the development of anthelmintic resistance” which was awarded the Australian College of Veterinary Scientists prize for the best scientific paper published in the New Zealand Veterinary Journal. Also, Dave Saville's 1990 paper entitled “Multiple Comparison Procedures - The Practical Solution” has recently been listed as the 15th most cited paper in a list of the 200 most cited papers in the 60 year history of the The American Statistician.

On the conference scene, Zaneta Park-Ng, Manasa Ramakrishna, Rajiv Chaturvedi and Anar Khan presented work at the 6th International Endophyte Symposium in Christchurch in March. Tanya Soboleva and Dave Saville were invited participants at a week-long weeds workshop at Hanmer in April, as part of a New Zealand Institute of Mathematics and its Applications (NZIMA) program on “Modelling invasive species and weed impact”. Ken Dodds attended the SmartGene for Beef workshop in Brisbane in May to discuss the integration of marker test information into genetic evaluation systems. Roger Littlejohn, Harold Henderson and Dave Saville attended the Statistical Association Conference in Christchurch in July, with Roger talking on “Hidden Markov models for feeding data from groups of red deer”.

Dave Saville

Department of Conservation
Statistical staff at DOC remain at two but with different faces. Maheswaran Rohan joined the department in November last year based in Hamilton, after Adam Smith left to take up a position at NIWA, Wellington.
At present, we are really busy running workshops and courses for DOC staff. Richard Duncan, Maheswaran Rohan, and Ian Westbrooke have conducted a three day workshop on analysis of repeated measures with R in Christchurch, which will be run again at the end of October. In the meantime we will be conducting statistical modelling courses in Christchurch, Wellington and Dunedin, and Dr John Leathwick from NIWA will be giving a two day workshop on boosted regression trees.

Rohan gave a talk on analysis repeated measures at DOC's Winter Science Convention, in Wellington in June. In July, Ian Westbrooke organised and gave a talk at the NZSA conference education session, and presented a paper on public favourability towards DOC at the Conserv-vision Conference, Hamilton. Ian then took time out to spend time with his mother, up to her death on 20 July.

Ian will be giving an invited keynote talk at Palmerston North Stats Forum in October, and is planning to run the workshop on regression at the Mathematics Teachers’ Conference in Auckland in September. Ian and Rohan enjoyed catching up with many in the maths/stats communities at the NZIMA Modelling Invasive Species and Weed Impact workshop at Hanmer, and the NZSA conference in Christchurch.

Maheswaran Rohan

University of Otago, Preventive and Social Medicine
The biostatisticians at the University of Otago's Department of Preventive and Social Medicine are slightly short handed at the moment as Katrina Sharples is on study leave for a year in Cambridge, working with Mike Kenward, and Peter Herbison is on six months study leave. He has just returned from working with colleagues at the Australasian Cochrane Centre, Monash University. Vicki Livingstone will be leaving us to go back to Ireland in 2008, and we are actively looking for a replacement at the lecturer/senior lecturer position. Melanie Bell has been working with the Department of Mathematics and Statistics developing a proposal for a Masters of Science in Biostatistics to begin in 2009. This joint divisional programme (Science and Health Science) will largely be a taught masters degree, with workplace consulting experience which she may ask for help with from her colleagues around New Zealand! If you are interested in hearing more give her an email (melanie.bell@stonebow.otago.ac.nz). Also, Melanie and Katrina are working with Stephen Vanderhorne in Auckland to organize a two day workshop on longitudinal data analysis, probably in February. More details to come.

Melanie Bell

University of Auckland
Our completing PhD students have had several major successes recently. James Russell came second in the 2007 MacDiarmid Young Scientist of the Year Competition, and was overall winner in the category Understanding Planet Earth. The result was announced at a grand awards dinner in Auckland on 20th June, and James’s research was discussed on TVOne News, Breakfast Programme, and Nine to Noon with Kathryn Ryan.

Meanwhile, Richard Umstaetter won a University of Auckland Best Doctoral Thesis award: one of five awarded by the university for 2006. Richard has now commenced his postdoc at the NASA Jet Propulsion Laboratory at Caltech. Christian Roever, who (like Richard) has just completed his PhD on Bayesian astrophysics with Renate Meyer, won a major student prize at the Joint Statistical Meetings in Salt Lake City, in the Bayesian Statistical Science section.

We extend a warm welcome to three new staff in the department: Stephane Guindon, Ivan Kojadinovic, and Mark Holmes. Stephane is a bioinformatics specialist from CRNS in Montpellier, France; Ivan is an OR specialist from the University of Nantes, France; and Mark is an Auckland graduate who is completing his postdoc in probability theory at EURANDOM. Welcome also to four new PhD students: Assad Ali (working with Renate Meyer), Bobby Wilcox (working with Alan Lee), Sohail Chand (working with Patricia Metcalf), and Irene Zeng (working with Sharon Browning).

Patricia Metcalf is involved in a $150,000 grant from the HRC for studies relating asthma and diet, and Alan Lee has gained a $70,000 OSResearch grant from Statistics New Zealand. Paul Murrell ran two R courses using an online discussion board: “Introduction to R” and “Graphics in R”, each four weeks long. Stephanie Budgett has been invited to give a summer course on “An Introduction to Statistics” by the New Zealand Social Science Network, and Andrew Balemi drew 35 attendees at an in-house Statistics New Zealand course with the title: “How I Stopped Worrying and Fell in Love with the Survey Cycle”. Andrew's winning aphrodisiac has not yet been revealed.

Finally, the Impact Factor of the Australian and New Zealand Journal of Statistics rose from 0.387 to 0.537 from 2004 to 2005, as a result of one paper by Marti Anderson. The publishers flagged this as a significant achievement for the journal, quoting in their recent report that: “The reason for this success ... is the Anderson article from 2003 which has received 29 citations”. Never one to sit back, Marti has recently settled a new contract with the US agency NOAA (National Oceanic & Atmospheric Administration) on analysis of habitat restoration projects.

Rachel Fewster

University of Otago, Maths and Stats
John Harraway attended ICOTS 7 in Salvador Bahia, Brazil in July. The conference was very worthwhile with 550 registered but about 15 not able to attend because of the collapse of the Brazilian Airline, Varig. A further 100 high school teachers from Bahia attended workshops in the two days before the conference. John presented a paper on Item Response Theory, took part in an invited debate about the teaching of statistics in context and organised the session on multivariate statistics within the tertiary education topic. He was Scientific Secretary for ICOTS 7 and has been asked to Chair the International Programme Committee for ICOTS 8 which will take place in Slovenia in 2010. John also visited the Federal University of Santa Catarina in Florianopolis where he is working with Dalton Andrade on a project in Brazil.

David Fletcher presented a paper on “Confidence Intervals for Expected Abundance of a Rare Species” at the NZSA meeting in Christchurch in July. This is joint work with Malcolm Faddy (Queensland University of Technology) and will appear in JABES in September 2007.

Mathew Schofield, Janine Wright, David Fletcher and John Harraway all presented papers at the NZSA Conference in Christchurch. Matthew Schofield shared the prize for best student presentation for his talk: Climate Reconstruction. In his talk Matthew highlighted problems in existing methods used for modeling climate from proxies such as tree rings and outlined how this should be done properly.

Irene Goodwin

Victoria University
Arrivals and departures are always big news and we have a few comings and goings to report since the last Newsletter. Colleen Kelly completed her two year term as the University’s Consulting Statistician in March 2007. Colleen will be missed very much by all of us and by the many staff and graduate students around the University that she assisted over the last two years. Also, Colleen and her husband, Cliff, will be missed in the Wellington cycling scene, where they had become pretty well known. Colleen and Cliff have returned to southern California (near San Diego), where Colleen has now started work for a private firm, following several years spent in academia. We wish them both all the best. Our group is now looking forward to a new arrival: the next Consulting Statistician is due to start work at VUW later this year, so there’ll be more on that in the next Newsletter.

Another departure this year was of Junko Murakami, who completed her postdoctoral position that was funded by the NZIMA programme on Hidden Markov Models, coordinated by David Vere-Jones. Junko is now in the US, but she wrote that her impression of New Zealand, after living here, is “something like a ‘tiny giant’. Incredible in many aspects.” She will miss NZ and the blue Wellington bay, and we will miss Junko too.

While not directly linked to the Stats and OR group, Sharleen Forbes joined VUW this year as Adjunct Professor in Official Statistics, in the School of Government. This is a half time position, shared with Statistics New Zealand. Sharleen is deservedly well known throughout the NZ statistical community, and her new professorial role is to promote official statistics: we wish her well with that. In fact she’s got off to a pretty good start, since Sharleen is organising an Invited Paper Meeting and speaking in a contributed Paper Meeting at the 56th International Statistical Institute in Lisbon, Portugal at the end of August; both meetings focus on teaching/education and official statistics.

In news from our research students, Nuovella Williams completed her PhD (“Robust Methods for Analysing Quantitative Trait Loci”), supervised by Richard Arnold and Ross Renner. Nuovella is now working for the Office for National Statistics in the UK. Two of our Masters students also recently submitted their MSc theses: Jason O’Sullivan (“A Taxonomy of Seasonal Patterns”) supervised by John Haywood, and Sarah Marshall (“Analysis of Reliability Data”), supervised by Stefanka Chukova. Jason is now working in London for dunnhumby, a Relevance Marketing company, while Sarah is heading off to the University of Edinburgh to start a PhD with full financial support. Clearly the UK is an attractive destination for our graduate students and we hope they all do well and enjoy themselves over there.

Sabbaticals are also prominent in the current news from VUW. Dong Wang returned from his travels at the end of April 2007, and restarted teaching immediately (welcome home Dong!). John Haywood continues to enjoy himself living in Santa Monica and visiting the UCLA Department of Statistics. John travelled to Europe in June 2007, where he presented a paper at the 3rd International Workshop on Correlated Data Modelling in Limerick, Ireland and also briefly visited Lancaster University, England, to help progress some work with Granville Tunnicliffe Wilson. In contrast to the extremely dry year that southern California (and John) is experiencing, it rained almost every day while John was in Europe. John now has a newfound appreciation for Colleen Kelly’s somewhat negative view of Wellington’s weather (Colleen is a SoCal native). In September John will present some research at the 2007 NBER/NSF Time Series Conference, which this year is to be held in Iowa City.

Richard Arnold began his sabbatical at Waseda University, Tokyo, visiting our former colleague Yu Hayakawa, who has been at Waseda since early 2004. Estate Khmaladze also started a sabbatical at the same time as Richard. Estate will travel quite widely, but initially he was based in Tbilisi, Georgia, where he reports it was very hot indeed. In Estate’s absence, back at VUW Yuichi Hirose has taken over the organisation of the group’s seminars. Yuichi also organises a separate stochastic process study group meeting each week.

Prof Michael Trick (Carnegie Mellon University), the 2007 ORSNZ Visiting Lecturer, visited the group (9-10 August) and gave two seminars: “Integer and Constraint Programming Approaches to Sports Scheduling” and “The Science of Better: Practical Operations Research”. Mark Johnston reports that the first talk was a fascinating insight into the computationally extremely-difficult problem of scheduling Major League Baseball in the USA, while the second talk outlined how Operations Research lies at the heart of businesses such as Google, FedEx and Amazon.

Some bad news concerned the free half-day workshop that Ivy Liu had organised on “Bayesian Analysis of Case-Control Data: Studies of Gene-Environment Interaction”, which was to be given at VUW in July 2007 by Bhramar Mukherjee, University of Michigan. More than 40 people had registered for the workshop but unfortunately it had to be cancelled, due to certain issues with New Zealand Immigration. The organisers extend their deepest apologies to all those affected. At the end of August Ivy will be presenting a paper at the 56th International Statistical Institute in Lisbon, Portugal, as will Dong Wang.
Shirley Pledger had a week of research meetings in July at the Max Planck Institute for Demographic Research in Rostock, Germany, followed by an invited speaker slot at a conference on Recent Developments in Capture-Recapture Methods and their Applications, at the University of Reading, England. En route she visited Goettingen, where Walter Zucchini took her to see Gauss’s grave. Shirley and the rest of us will see Walter Zucchini at VUW next year, since he will be the 2008 recipient of the Shayle Searle Visiting Fellowship in Statistics.

John Haywood

Wellington Statistics Group
The Wellington Statistics Group (WSG), a local group of the NZSA, continues to meet regularly. Since the last NZSA Newsletter there were WSG talks given by:

6 August: Nick Longford, SNTL, Reading, UK, “Allocation of limited resources and related problems in small-area statistics”.

9 May: Shirley Pledger, Victoria University of Wellington, “Something for Nothing: Estimating age-related survival rates from capture-recapture data when age is unknown”.

18 April: Stephen E. Fienberg, Carnegie Mellon University, USA, “When Did Bayesian Inference Become “Bayesian”?”

22 March: Richard Arnold, Victoria University of Wellington, “Using earthquakes to measure stress in the earth’s crust”.

Abstracts and further details of past and known future WSG talks can be found on the NZSA Local Groups web page. That web page also contains contact details for the group and information on how to subscribe (subscription is free) or how to update email contact details. WSG gratefully receives regular sponsorship from the Ministry of Social Development, Statistics New Zealand, Statistics Research Associates Ltd, and Victoria University of Wellington.

John Haywood

University of Waikato
It has been a while since there has been any news contribution from the Statistics Department at the Waikato University. That is rather symptomatic of the ways things are here. As we are short-staffed (no replacement yet for James Curran or Nye John), we have all been busy with our University commitments.
As for recent activities in the department, we welcomed Lyn Hunt back after illness. She is still having to take things easy but did manage a visit to Brisbane in early June to work with Kaye Basford. The department hosted Professor William Reed from the University of Victoria, Canada ,who was here to work with Murray Jorgensen. Murray also went to Barcelona in July to the International Workshop on Statistical Modelling 2007 where he presented a paper entitled “Multivariate Mixture Models in official statistics”. At the same time, Judi McWhirter attended the NZSA 2007 conference in Christchurch where she presented a paper entitled “Who has mud on their hands? A bootstrapping technique for determining a fingerprint for sediment tracing in the Whangapoua Harbour”. The second edition of Bill Bolstad’s book “Introduction to Bayesian Statistics” has just been released.

The department ran another successful workshop in April. Professor Adrian Baddeley, from the University of Western Australia, presented a Spatial Statistics Workshop, which was well attended. Waikato University has been decided as the venue for the next NZSA conference. It is likely to be held in early July 2008, but dates are still to be confirmed.

Recent seminars in the department:

Murray Jorgensen (University of Waikato) “Where have all the Young Men gone?”

Bruce Weir (University of Washington) “The Birthday Problem and DNA Profiles.”

William Reed (University of Victoria) “Normal-Laplace Distributions and their Applications”

Hadley Wickham (Iowa State University) “Interactive Graphics.”

John Leathwick (NIWA) and Jane Elith (University of Melbourne) “Boosted Regression Trees - a major advance in statistical modelling tools.”

Judi McWhirter

Statistics New Zealand
As usual I’ll start with arrivals and departures to Statistical Methods since March 2006, so many of you can find out where your students have ended up. Arrivals: Jing You, Rudi van der Mescht, Emma Mawby, Deborah Brunning, Olivia Son, Guan Yu (Fish) Zheng, Lisa Henley, Lilian Morrison, Chang (Lisa) Li, Kee Chang, Chris Hansen and Val Cox. Returning: Frances Krsinich and Diane Ramsay. Departures: Sue Brown (Timaru) Jo Shrigley (Australia), Polly Stuart (Qatar), Joh Prebble, Zoe Wood (OE), Karla Helgason (StatCan), Steve Johnston, Eleanor Posadas (MSD), Katrina Young (Tourism), Tim Duke, Alan Bentley, Emma Hooper (ONS), Jeremy Wilhelm (ACC), Simon Leong (NZER), Andy Smith (SSC) and Tiri Sullivan (Emirate Airways). There have also been people moving to and from Statistical Methods to and from other areas of Statistics NZ which I haven’t logged. All this shows that I need to report more regularly to keep the list to a manageable size!

In other news, Jamas Enright has recently been seconded to Treasury for 2 years. There has been the welcome return of Diane Ramsay as Manager Statistical Methods. She’s just had the joy of going through the 50 odd (as in approximately, not a comment of the personalities of Methodologists in Statistics New Zealand) performance reviews, and still seems cheerful. Victoria Wei had a baby girl in February, and Claire Sun and Sela Appleton have recently gone on parental leave to await their happy events. In March we had colleagues from Tonga, Palau and Samoa here to upskill themselves on various aspects of official statistics, and many from Statistics NZ assisted in this. The Pasifika statisticians very much appreciated the opportunity, though not Christchurch’s weather. We also hosted Stephen Fienberg for a week prior to his visit to VUW. Steve has been doing confidentiality work in association with Fraser Jackson and us as well as for the U.S. official statistics system, so it was an excellent opportunity to talk confidentiality (confidentially of course).

We have also been a peripatetic lot. Rebecca Bangma, Sela Appleton and Claire Sun went to the Australian Young Statisticians conference in Melbourne. John Crequer and Frances Krsinich are off to the ISI meeting in Lisbon. John is talking about our use of the electronic transactions data (EFTPOS data to you and me). He also is taking the opportunity to visit various researchers in Europe to discuss temporal disaggregation and seasonal adjustment issues. John Pearson went to the 3rd International Conference on Establishment Surveys in Montreal where he presented a paper on our programme for generic survey systems. He also took the opportunity to visit Statistics Canada. We sent Walt Davis, Mike Camden,Tim Duke and Jing You to MISG07 at Wollongong. Local conferences with SM attendees and presenters included Social Policy Research & Evaluation, Population Association of New Zealand and of course the recent NZSA conference, which Rebecca Bangma also assisted in organising.

Recent Statistics New Zealand news include a new Government Statistician, Geoff Bascand. Also we have received some money from government to enable us to make much more of our data available free. The projects approved for funding by OS Research programme have been decided. They cover confidentiality, organising administrative data, quality assessment of linked data and sampling the Maori and Pacific populations. However I need to stress that relying on my report to know what work is being done in Statistics NZ is much, much less useful than going to www.stats.govt.nz!

Richard Penny

Massey University, Turitea
We welcome two new graduate students: Nafees Ahmad has arrived from Pakistan, Sarojinie Fernando has arrived from Sri Lanka. Sigrid Panisch from Austria will come later in the year.
Chin Diew Lai attended and presented a paper at the ISSAT International Conference on ‘Modeling of Complex Systems and Environments’, Ho Chi Minh City, Vietnam, 16-18, July, 2007.

Around Easter, Doug Stirling had another visit to Nestlé in Switzerland and the University of Reading in England, finishing one CAST project and starting work on another. He also released a new version of CAST with improvements to the basic e-books and extra e-books about multiple regression and design of experiments.

A large contingent from Palmy attended and presented at NZSA2007. Wes Johnson visited briefly on his way home to work with Geoff Jones.

Another Palmy Statisticians Forum is being planned for 26th October.

Geoff Jones

Return to top