AgResearch - John Koolaard
Auckland University of Technology - Jeffrey Hunter
Canterbury Young Statisticians - Richard Penny
Department of Conservation - Ian Westbrooke and Maheswaran Rohan
Massey University, Albany Campus - Marie Fitch
Massey University, Manawatu Campus - Geoff Jones
Plant and Food Research - Duncan Hedderley
Statistics New Zealand - Richard Penny
Statistics Research Associates - David Harte
Victoria University of Wellington - John Haywood
University of Auckland - Rachel Fewster
University of Canterbury - Peter Smith
University of Otago - Austina Clark

AgResearch

John Koolaard

Auckland University of Technology

Jeffrey Hunter

Canterbury Young Statisticians

Canterbury Young Statisticians at their mixing event.

Richard Penny

Department of Conservation

Ian’s office post 7.1 earthquake in Christchurch.

The 7.1 earthquake threw the filing cabinet drawers open, and spilled some contents. A colleague was heard to comment that Ian’s work area looked pretty much as usual. But no serious damage at work or home. Some DOC colleagues homes were less fortunate - with the main issue being damaged chimneys. Ian managed to miss the earthquake by a day, by taking an extended tour after the International Conference on Teaching Statistics in Slovenia, where he presented in a session organised by Jennifer Brown on workplace training in environmental statistics. Europe was sweltering in a heat wave, with Ljubljana reaching 35C, but fortunately the conference venue was air-conditioned. There were plenty of New Zealanders enjoying the warmth at ICOTS, and at the International Statistical Ecology conference in Kent the previous week.

Now it is time for Rohan and Ian to catch up on training work at DOC before staff become too absorbed with field work. With the assistance of Tim Robinson (Uni of Wyoming) they piloted a new three day course on Designing Studies in May. It will be repeated twice in October, and there is a lot of interest for next year. There is also continued interest in courses on linear and generalised linear models using R, and matrix population modelling.

Rohan and Ian enjoyed catching up with people at the NZSA conference in Palmerston North. They presented on their positive experiences introducing R using the R Commander Menu interface. There is substantial interest in this and they have made similar presentations at University of Otago Christchurch, and at Landcare Research, Hamilton.

In March, Marti Anderson hosted a one-week workshop on multivariate analysis for biologists and ecologists, the second to be held here at Massey Albany, with a focus on the computer software PERMANOVA+, which was developed in collaboration with colleagues at Plymouth Marine Lab (PML) in the UK. Similar workshops were also presented by Marti in August in Seattle, Washington, for the National Oceanic and Atmospheric Administration (NOAA) and in June in Plymouth at the home of the Marine Biological Association (MBA) of the UK. The MBA lab looks right out over Plymouth harbour, where Sir Francis Drake defeated the Spanish Armada. Right down the lane are also the stone steps down to the water where Marti’s direct ancestor, William Bradford, author of the Mayflower Compact, embarked with his fellow pilgrims to the New World. What goes around comes around, I guess! Marti also took part in a workshop for the analysis of beta diversity at the National Centre for Ecological Analysis and Synthesis (NCEAS) in Santa Barbara in April. This was the second of three sessions devoted to the topic, with the next one to be held in November.

The other exciting news for Marti, post-doc Dr Mat Pawley, and PhD students Adam Smith and Oliver Hannaford, was the delivery of a brand new 4.3m Naiad rigid inflatable boat (RIB), which is just perfect for diving and counting fish (among other things). While the crew are busy getting together all of the materials for the boat (the list is long - everything from lifejackets to scuba tanks) and brushing up on boating skills, first aid, and other qualifications, an important decision had to be made - what shall we call it? We settled on the name “RV Poisson” (the “RV” bit stands for “Research Vessel”), which seemed only fitting for a craft specially designed to serve marine ecological statisticians!

In August Beatrix Jones hosted Anthony Fiumera from the Biology department at Binghamton University in Binghamton, New York. They are working together on a project to detect genetic variants affecting reproductive success in wild populations of fruit flies (Drosophila melanogaster).

Adam Smith successfully completed his PhD confirmation. Well done Adam! In June, Marie Fitch presented some of her PhD research at the ‘Sparse structures: statistical theory and practice, Research workshop’ in Bristol, UK and the WNAR/IMS annual meeting in Seattle, Washington. Aside from the expected relevant presentations and networking opportunities, a highlight was the Bristol conference dinner which was held on board the SS Great Britain.

One of the team members, Dr Howard Edwards, has moved from his old office in the Quad building to an office in the IIMS building with the rest of us, so our group is now all housed in the same place at last. Howard, Beatrix Jones and Marie Fitch have been beavering away on material for an online textbook resource for the stage 1 Business Statistics paper. This project is being co-ordinated by colleagues in Palmerston North and also involves those teaching the paper in Wellington. Marie and student Katharina Parry have also been doing some liaison work with presentations to local schools. Meanwhile John Xie, another PhD student, has engaged in more family-based liaison supporting his son David who was a member of the team of secondary school students who brought a gold medal home from the 23rd International Young Physicists’ tournament in Austria. (Well done.)

Finally, a recent important discovery was the ping pong table at Ferguson’s, the local pub right here on the Massey Albany campus. Dr Mat Pawley is currently the reigning table tennis champion in the statistics group.

Our last newsletter entry heralded the imminent arrival of a firstborn to Jonathan Godfrey and wife Olivia. Callum Joseph Godfrey finally saw the light of day at 2.25pm on Sunday 11 April 2010, having capriciously missed the newsletter deadline. Young Callum tipped the scales at 8lbs 9oz. More recently, Jonathan has been in the news for his Judo exploits (stuff.co.nz article).

Two of our postgraduates have completed their PhDs. Ting Wang, Mark Bebbington’s student, has had her thesis on “Hidden Markov Models for Geophysical Hazards” accepted (and added to the Dean’s list). Ting is still at Massey, having taken up a postdoctoral research position at Volcanic Research Solutions. She is currently in Europe to attend a vulcanology workshop in Italy. Maris Isidro, Steve Haslett’s student, is just completing the emendations to her thesis on “Intercensal Updating of Small Area Estimates”. Maris has just started an internship at the Ontario Ministry of Economic Development and Trade, having moved to Canada to join her husband.

Alasdair Noble is recovering from his leading role in organizing a very successful NZSA Conference (report here), just in time to help Ganes organize the next Palmy Stats meeting in October (to see the programme click here).

Congratulations to Steve Haslett who was presented with the Campbell Award at the NZSA Conference Dinner [article here]. Steve has been travelling a great deal recently – to China as an invited speaker at the International Workshop of Matrices in Statistics, to Canada for the Joint Statistical Meetings, to Cambodia for a poverty mapping assessment for the World Food Programme, to Kiribati and Vanuatu for a UNICEF education project, and to Thailand for an NZAID review.

Martin Hazelton attended the Joint Statistical Meetings in Vancouver in early August to give a talk on “Analysis and Modeling of Networks”. He arrived back in New Zealand in the wee small hours and, showing true dedication, rushed in to Massey to give his morning lecture. More recently, Martin gave the second in the Professorial Lecture series recently instigated by the Institute of Fundamental Sciences to showcase the Institute’s research to others. His lecture on “Change and Changeability: Some Modern Directions in Statistics (with apologies to Jane Austen)” drew an unexpectedly large audience. Whether they were primarily fans of Statistics or Jane Austen has not been determined.

Doug Stirling has also been building up his frequent flyer points this year with visits to Slovenia (ICOTS), Belgium (ENBIS conference) and Canberra (ANU), and has another trip to Perth (OZCOTS) planned for December. He has also been developing CAST-based resources for NZ high schools to help teach NCEA statistics. How he finds time to do this is a mystery, as he’s also being bombarded with requests by the Business Statistics team who are now using CAST for their first year course.

Geoff Jones attended the 25th International Workshop on Statistical Modelling in early July. This being his first visit to Glasgow, he was quite surprised to be awakened on his first night by the sound of breaking bottles and loud shouts in an unintelligible dialect. It did stop raining at one point, giving the delegates a chance to explore the beautiful city. Geoff’s presentation on “A Longitudinal Model for Multiple Diagnostic Tests” was scheduled as the first talk on the morning after the Conference Dinner, but nevertheless a surprising number of people struggled in to listen to it.

Finally, our re-branding exercise continues: extramural students are now to be called “distance students”. No doubt redundancy will soon be rebranded as “extended leave”.

Our biggest news is no news – we are still looking for a team leader. Peter Alspach has agreed to be acting leader until the end of the year.

One of the first things Peter organised was a team meeting in Palmerston North at the end of June. There was a good exchange of experiences – Esther Meenken talked about mixed models and soils, Andrew Wallace about insect trapping experiments in Perth, Mark Wohlers about assessing how good people’s sense of smell is. Andrew McLachlan and Marcus Davy acquainted us with wikis, sharepoint, subversion and other collaboration tools.

It was a busy week for Andrew McLachlan – the day after the team meeting he talked to the young statisticians at the NZSA conference, and later that week gave a presentation of GenStat for Schools to teachers.

Marcus Davy was an invited speaker at the New Zealand Next Generation Sequencing Conference and presented a seminar on quality control issues in next-generation sequencing experiments. This was a bit of a hot topic with half of the talks discussing issues surrounding the technology. He also attended the BestGRID winter retreat and presented a seminar on developing and using GenePattern modules for microarray analysis and next-generation sequencing in a half hour session with discussion. This involved an interactive demonstration of developed modules using a GenePattern server on a laptop.

Mark Wohlers is nearing the end of the first year of a Master of Applied Statistics course with Massey University and is starting his thesis project on design and analysis of odour threshold determination experiments.

Esther Meenken will be attending the International Biometric Conference in Florianopolis, Brazil, in December.

I will start my report with our great pride in Mike Doherty and Mike Camden receiving honorary life membership of the NZSA at the recent NZSA conference. As we have no other Mikes working for us we see it being a few years before having anyone here receiving a similar honour. Our presence at the NZSA conference was large as usual with Chen Chen, Walter Davis, Hazel Kale, Rebecca McGirr, Tim Hawkes, Andrew Richens and Guan Yu (Fish) Zheng presenting talks. I refer you to the NZSA conference site for details on their 6 talks. Rebecca also filled in for Richard Penny as secretary for the AGM and got elected to the NZSA executive for her trouble. Richard was re-elected as secretary.

The recent Canterbury earthquake has made the third floor of the Christchurch office where Statistical Methods sits temporarily unusable. As a result we are scattered about the second floor, though we manage to still have group get togethers, such as for the daily answering of the Dominion Post quiz. Our weekly seminar series continues as well, recently covering such topics as aspects of the Household Labour Force Survey, the Time Use Survey, composite estimation, businesses’ nonresponse, and data validation methods. We were also fortunate to have Tommaso Proietti, an Erskine Fellow at Canterbury, present a talk on outliers in time series. This also covered seasonal breaks and we are hoping to work with Tommaso on this in the future. While Tommaso was available we took the opportunity to host a workshop on temporal disaggregation which gave us some good ideas for our work. Statistics New Zealand was also delighted to host the inaugural gathering of Canterbury Young Statisticians.

Statistical Methods recently had its annual professional development off-site in Wellington with 2 guests from the Australian Bureau of Statistics. Professional development included a session on how to be a good consultant. It’s somewhat frightening how good most people were at being stroppy clients in the role-playing exercise. Perhaps we’re just good at channelling the clients we’ve had in the past. While technical issues were covered we also had a debate on “Let the Data Speak”, in which the 6 debaters were somewhat at variance as to what it meant. Penny Barber provided the highlight when she showed us that we were all evil statisticians! We also recently hosted a group of statisticians from the Korean statistical office which demonstrated the diversity of our employees as Soon Song, who used to work in the Korean office, and Olivia Son were available to provide translation of the technical discussions.

With the upcoming Population Census in March 2011 and subsequent surveys, such as disability, based on the data from the Census soon after, the people working on those are seeing the end in sight, though of course this means there is a lot to do. And as data collection for official statistics is like painting a large bridge there is already work under way on the shape of the data collection for 2016 Census.

Given the long time since my last report I need to mention the election to the International Statistical Institute of Christine Bycroft. She has also organised an invited session for next year’s ISI meeting in Dublin and will be presenting a paper at the Statistics Canada annual methodology symposium in October.

As there are so many I won’t report on the comings and goings since my last report except for the recent departure of Harry Smith after almost 25 years in Statistical Methods, which had him as our 3rd longest serving member.

The main news from SRA is that recently the SRA directors unanimously elected David Vere-Jones to the position of SRA Distinguished Associate in recognition of his far-sighted and seminal contributions to SRA over the last decade as foundation director, shareholder and associate of SRA. These personal contributions have been fundamental in establishing the strong reputation and standing of SRA as a high-quality national and international statistics research organisation.

Other news: Wang Ting submitted her PhD thesis “Statistical Models for Earthquakes Incorporating Ancillary Data” at the beginning of 2010. Her oral examination was held in May 2010 which she passed with flying colours! David Vere-Jones and David Harte were co-supervisors with Mark Bebbington at Massey University.

Robert Davies is updating his vehicle crash risk model for Opus International Consultants using an updated set of data.

Peter Thomson continued his involvement with NIWA (focused on weather generator models), the VUW School of Economics and Finance (adjunct professor), and Keio University, Japan, where he gave an invited paper “Towards a robust statistical framework for the assessment of quality of supply by New Zealand electricity networks” at a Mini Workshop on Leading Edge Data Analysis. He completed a report for the Electricity Commission that reviewed the methodology and models used by the Commission for national electricity demand forecasting, undertook consulting for the Department of Corrections, and presented the paper “A hidden seasonal switching model for high resolution breakpoint rainfall data” (joint with John Sansom) at an invited session of the fifth International Workshop on Applied Probability (IWAP 2010) held near Madrid.

David Harte was back at the Institute of Statistical Mathematics in Tokyo again during March, working in Yosi Ogata’s group on spatial ETAS point process models. He has also been involved in a joint OSR project with Ken Richardson and Kristie Carter at the Wellington School of Medicine applying hidden Markov Models to longitudinal data.

Both David Vere-Jones and David Harte attended the CORSSA workshop at ETH in Zurich in May. The attendees are developing web based resource material for statistical seismology. David Harte returned via Tokyo and gave a talk at a meeting on Statistical Seismology at the ISM. After the Zurich meeting, David Vere-Jones visited Katerina Orfanogiannaki at the Institute of Geodynamics in Athens, and also visited Professor Karlis at the University of Athens where he gave a seminar. On the way back to NZ he visited the Institute of Statistical Mathematics in Tokyo, and was one of the first guests to stay in Akaike House. It adjoins the new ISM in Tachikawa.

Much of this news entry concerns reports of people’s travels, along with great news of one new (unpaid) arrival: Mark Johnston and his wife Emily, along with big brother Hamish (actually, really big for someone who has recently turned two!) welcomed the safe arrival of Benjamin Johnston on 05/05/10; we already see the potential for some maths games with birth dates... Benjamin took only one hour to arrive following his mum’s check in at the hospital, so he’s already known for his speed out of the blocks. Congratulations to Mark and Emily! Between teaching and parental leave, Mark managed to fulfil his pre-planned conference visit to the 2010 IEEE World Congress on Computational Intelligence (18-23 July in Barcelona) where he was an author/presenter on two papers.

Mark’s co-authors included one of our Honours students, Thomas Liddle; Thomas quite recently got married, so congratulations to Thomas too!

In June and July Richard Arnold had an extended visit to Spain and the UK. Richard attended the Bayesian Valencia conference (the last of its kind: from now on ISBA will have a Bayesian meeting every two years, without returning to a particular location). Shirley Pledger and Richard presented a workshop and two papers at the International Statistical Ecology Conference at the University of Kent, UK (at which all four workshops were presented by New Zealanders!). The workshop was on capture-recapture models for open populations (with age structure and heterogeneity of capture probabilities), and the papers were on clustering of species and habitats in community ecology using finite mixture models. Also in June, Shirley gave a seminar at Murdoch University, Australia, on pattern-detection models in community ecology.

John Haywood organised the Wellington leg of the New Zealand Statistical Association Visiting Lecturer tour in early July. This year the NZSA Visiting Lecturer was Prof Ingram Olkin (Stanford University), whose visit was associated with the NZSA / Statistical Methodologies Conference, 29 June-1 July 2010 at Massey University, Palmerston North (see http://nzsa_cdl_2010.massey.ac.nz/). John also presented a paper at the NZSA Conference and then brought Ingram down to Wellington on 1 July. Prof Olkin gave four talks while in Wellington: a Wellington Statistics Group talk on 2 July (Meta-Analysis: History and statistical issues for combining the results of independent studies; also see below), a repeat of that first talk to the Ministry of Education on the morning of 5 July, an afternoon talk on 5 July to Statistics New Zealand (Measures of heterogeneity, diversity and inequality), and a lunch time seminar at Victoria University of Wellington on 6 July (Life distributions in reliability and survival analysis). See the attached photo of John and Ingram taking a breather while discussing wavelets (or something similar...) at Lake Ferry on 4 July, during Ingram’s Wellington visit. Some details of the rest of Ingram’s NZ tour are here:
http://stats.org.nz/visiting_lecturer_2010.shtml.

Estate Khmaladze and Ivy Liu both attended the Joint Statistical Meetings in Vancouver (31 July to 5 August). Ivy organized and chaired a session on Novel Methods for Extended Case-Control Designs, while Estate presented some work in a session on Model Diagnostics. Besides his JSM session, Estate also visited nine other countries on a recent two-month trip overseas (finishing with the JSM), during which he gave a total of 14 presentations in a very busy schedule. Stefanka Chukova also spent time abroad recently, including a visit to Bulgaria between our trimesters. Fortunately Stefanka came back to continue leading the preparation for APARM 2010 this coming December, which is shaping up nicely (see http://msor.victoria.ac.nz/Events/APARM2010/).

Estate and Ivy are also involved in the organisation of two further events that we are hosting towards the end of the year, but both preceding APARM 2010. First, along with Dong Wang, Ivy is organising the visit of Prof Kai-Tai Fang (Chinese Academy of Sciences and BNU-HKBU United International College). Prof Fang will be the 2010 Shayle Searle Visiting Fellow in Statistics at Victoria University; he will be here in New Zealand from 19 November to 4 December. During his visit, on Tuesday 23 November, Prof Fang will deliver a free one-day short course, entitled “Computer Experiments: Design and Modeling Workshop”. Further details are available at the course web page, http://msor.victoria.ac.nz/Events/ShortCourse2010.

Secondly, Estate Khmaladze is the main driving force behind the Second Wellington Workshop in Probability and Mathematical Statistics (30 November and 1 December 2010), which follows on from the first such workshop that ran successfully in November 2009. Helping Estate on the Programme Committee this year are David Vere-Jones and Ilze Ziedins (University of Auckland). Attendance at the workshop is free, and expressions of interest (along with titles of proposed presentations) should be sent to Estate Khmaladze or Ilze Ziedins by 31 October 2010. Further details are available at the workshop web page, http://msor.victoria.ac.nz/Events/WWPMS2010/, which will be regularly updated with any relevant new information.

Some brief but very good news concerning our students is that Giorgi Kvizhinadze (supervised by Estate Khmaladze) successfully defended his PhD entitled “Large number of rare events: Diversity analysis in multiple choice questionnaires and related topics”. Congratulations to Giorgi!

Wellington Statistics Group

Finally, in news from the Wellington Statistics Group, John Haywood has resumed as WSG convenor, replacing David Harte who has been the convenor since 2007. Recently David has been out of New Zealand for some lengthy periods and he felt it was time to step down from the role. John was previously convenor from 2001 until 2006. Alistair Gray continues to be the treasurer and Leigh Roberts usually organises the refreshments, while John maintains the group’s mailing list. WSG talks since the last NZSA newsletter were given by:
26 May 2010, Shirley Pledger and Richard Arnold (VUW), “Clustering and Pattern Detection in Ecological Models Using Mixtures”
2 July 2010, Ingram Olkin (Stanford University, NZSA Visiting Lecturer), “Meta-Analysis: History and Statistical Issues for Combining the Results of Independent Studies”

Further details of these talks can be found on the NZSA Local Groups web page: http://stats.org.nz/local_groups.shtml. This web page also contains contact details for the group (including the mailing list web interface), names of sponsors, and details of forthcoming (or most recent) talks.

John Haywood

University of Auckland

Rachel Fewster

University of Canterbury

Peter Smith

University of Otago

Austina Clark