Download PDF

Download this issue as a PDF.

President’s Corner

Portrait of Elizabeth ThompsonGreetings!

It seems no time since, with some trepidation, I sat down to write my first President’s Corner, and now this one is my last, although there is still much to do in the next two months, before I hand over the baton to the incoming president Louise Ryan.  This year has seen many regional IBS meetings, and next year will see IBC2018 in Barcelona.   There is much happening in the world, but our meetings are still places where people from so many countries come together to share our common interest in applications of mathematics and statistics across the biosciences.

We are a Society of Regions. And new regions, as well as the continuing success of established regions, are the key to our future. In early 2016, my presidency started with the Board’s approval of the new Malawi Region, which has continued to thrive.  Malawi has just hosted a very successful SUSAN meeting, and is taking over leadership in SUSAN.  In October 2017, at my last Board meeting as president we approved two new regions in Africa:  the North Africa Region, currently representing members in Morocco, Algeria and Tunisia, and also a new Tanzanian Region.  A big thank you to long-standing IBS members who helped to bring these new regions into being, and a very warm welcome to the members of these new regions as they join the IBS family.

As you all know, if you follow IBS news at all, we had a great 70th birthday celebration on September 6th, and thanks go to IBO and Wiley staff, as well as many IBS members. The web and email birthday announcements went (almost) without a hitch, and were very well received.  We had concurrent brief editorials in JABES and Biometrics: thank you to Marie Davidian (Biometrics) and Steve Buckland (JABES).  We have a wonderful video by Katya Ruggiero (AR):  this is available for all who would like to use it at their meetings.   Also, Biometrics has a facelift – although the word “facelift” in the announcement header sent the email to spam for many IBS members.  For the year, we have a new 70th anniversary IBS logo on the journal and for use in other anniversary contexts. The new rotating cover pictures and the new cover will continue and bring a new more exciting look to our Journal.   Very many thanks to Louise Ryan, Marie Davidian, and Steven Ottogalli (Wiley) and many others for seeing this project through.

This is also the time to thank so many who have helped to make these last two years so successful for IBS, and so enjoyable, although sometimes challenging, for me.  For reasons of space, I will not include those whose contributions are acknowledged in recent President’s Corner pieces, and I apologize in advance to any others I inadvertently omit.

Thank you to Executive Director Peter Doherty and all the staff at the IBO, where we have seen major transitions over the last two years.  Thank you to John Hinde and James Carpenter who smoothed the transitions in my first year as President in 2016, and to Louise Ryan and Brad Biggerstaff for taking up the baton in 2017.   Thank you also to all members of the Executive Board for their timely input and good judgment.  IBS is indeed fortunate to have such dedicated Board members, who volunteer much time and thought to our Society.

A special thank you goes to the chairs of our Standing Committees, and through them to the committee members.  You will find a note thanking all members whose committee terms are expiring elsewhere in this Bulletin*.  It is in the Committees that the major ongoing work of the Society is done.  Thank you to Awards Fund Committee chair Dan Kajungu, who not only oversaw the IBC DC Travel awards in 2016, but who met the challenge of a hugely successful non-IBC year travel awards year in 2017, with 15 awards made to members from 9 different countries to attend 8 different IBS regional meetings.  Thank you to Finance and Budget Committee chair Rene Eijkemans: his committee keeps us on track financially, while also allowing for our enthusiasm to fund ad hoc budget increases to the Awards Fund and the Representative Council fund for Networks and Inter-regional Collaboration.

Thank you to Education Committee chair Pascale Tubert-Bitter.   With members of her committee, the IBS Journal Club has this year got off to a great start, with 5 excellent sessions between the launch in April and the end of 2017.  The Education Committee has also selected 5 Short Courses for IBC2018, from a strong list of 34 proposals, and has organized our now traditional Statistics-in-Practice IBC session.  Thank you to Conference Advisory Committee chair Andrew Zhou.  His committee is now undertaking the difficult task of looking further ahead to locations for the IBC’s of 2022 and 2024 – no easy task in these uncertain times.

Thank you to Editorial Committee chair Esa Laara, and to Communications Committee chair Katya Ruggiero:  these two committees played an important role in the process for changes to the face of Biometrics (see above). On the journals also, I thank Mike Daniels (ENAR) as he ends his service as Biometrics co-editor, and welcome Debashis Ghosh (ENAR) as he takes up that role.  I also welcome David Warton (AR) and Hans-Peter Piepho (DR) as the new IBS representatives on the JABES management committee.

Last, but far from least, I give special thanks to Andrea Berghold  for her long and distinguished service as chair of Representative Council (RC).  Our Board liaisons connect the Board to the Committees.  Our Committee chairs, and other RC members on committees, link the Committees to the Council.  But the role of connecting the Council with the Officers and Board falls heavily on the RC chair.   Serving as an ex officio non-voting member of the Board, Andrea has played a key role in establishing the role of the RC under the new governance system, and in bringing the regions closer to the centre of IBS, while still maintaining the balance of responsibilities and roles of the Board and RC.   Finally, she has established a new system (paralleling our schedule of incoming and outgoing Presidents) to insure more continuity and effective leadership of the RC.   So although Andrea steps down as RC Chair on 31 December, she will serve one final year as Outgoing RC Chair.  I therefore welcome her in that new advisory capacity, and welcome also her successor as RC chair.

No newsletter at this point in our biennial cycle would be complete without at least a brief look ahead to the IBC.   Broadly, thanks to many, but especially to IPC chair Charmaine Dean and LOC co-chairs Pere Puig and Lupe Gomez, all is on schedule.  The web site contains more information week-by-week.  The abstract system is open for submission of abstracts for both the invited papers and contributed paper and poster sessions.  The calls to Travel Award applicants, and to our Young Statisticians have gone out.   We expect Registration to open on time in December.

So now it is up to you:  please submit your abstracts and make your plans to visit Barcelona in July 2018.  I hope to see many of you there!

Elizabeth Thompson

From the Editor

Dear Readers,

I am happy to bring you the last issue of 2017. In this issue we publish an article on the missing data topic group (TG), a first in a series of papers on the Strengthening Analytical Thinking for Observational Studies (STRATOS) initiative. In each article we will give a general description of the TG, aims and a few examples that illustrate good statistical practice. I have a special interest in this TG since this has been the topic of my research since 2013.

Daria Steigman, our professional journalist, has written an article about Statistics and Food Security – how data analysis can help us feed the world, based on an interview with Prof Kay Basford (University of Queensland). In this article Prof Basford illustrated the role of statistics in plant yields and demonstrated bio-fortified foods, which are plants bred to survive in specific conditions and fortified with key nutrients that people need but don’t have.

In the Software Corner Rebecca Barter, a third-year PhD student (Department of Statistics, UC Berkeley US), wrote an article about her R package, superheat, that provides an extremely flexible and customizable platform for visualizing complex datasets.

Louise Ryan, IBS Incoming President, wrote an article related to her presentation at the Spring 2017 meeting of ENAR, entitled “But I’m a Data Scientist Too, Aren’t I”.  In general Louise refers to Statistics with large and complex data, and gives examples from some of her recent projects. She concludes that nowadays a good biometrician should have additional Computer Science skills beyond the blend of Mathematical/Statistical and Domain Knowledge.

In this issue we publish a solution to the mathematical riddle from the previous issue, including the names of the first five to answer correctly. We received only 5 correct answers out of 27 answers. The first one to answer correctly was: Michael McIsaac (Queen’s University, Ontario). I’ll appreciate if you could email me any interesting riddles you encounter to be published in a future issue.

We welcome the two new regions in Africa to the IBS family: North Africa Region and the Tanzanian Region.

I would like to thank Alphonsus Baggett for his excellent help with the Bulletin over the past 2 years while I have been the editor, and to welcome his successor Kristina Wolford.

The  Journal Club Session, organized by the Education Committee of IBS, was held on December 14th.It was centered around the paperHandling Missing Data in Matched Case-Control Studies Using Multiple Imputationby Shaun R. Seaman and Ruth H. Keogh. More details were announced on IBS website.

We encourage you to submit abstracts to the IBC2018 conference to be held on 8-13 July 2018 in Barcelona, Spain, by the 10th of January 2018

Havi Murad

Thank You to IBS Outgoing Committee Members

Often in the background, but very much a part of the IBS
governance structure, are our standing committees. Each committee is made up of volunteers from numerous IBS Regions. And while the background and personal geography of each volunteer may differ, it is the unique combination of viewpoints and expertise that is crucial to the successful analysis and completion of committee projects. Together, these volunteers have done an admirable job in representing the IBS membership in 2017, for four or for even eight years! At this time, the IBS would like to thank those committee members whose terms are ending 31 December, 2017 for their service to the Society:

Awards Fund Committee: Daohai Yu (ENAR), and Tomi Mori (WNAR)

Committee on Communications: Dilip Nath (Indian Region), Ethelbert C. Nduka (Nigerian Region), and Raul Macchiavelli (Central American-Caribbean Region)

Conference Advisory Committee: Carolyn Rutter (WNAR), Francesca Little (South Africa), Gheorghe Luta (ENAR), and Herbert Thijs (Belgian Region)

Editorial Advisory Committee: Julio M. Singer (Brazilian Region), Ming Tan (ENAR), and T. Shun Sato (Japanese Region)

Education Committee: Yehenew Getachew (Ethiopian Region)

Budget & Finance Committee: Dipankar Bandyopadhyay (ENAR), Katja Ickstadt (German Region), and Victor Leiva (Chilean Region)

By Peter Doherty

XXIX International Biometric Conference

Young Statisticians Showcase

The Young Statisticians’ Showcase is an important part of the Society’s International Biometric Conference (IBC). If you are currently an MSc or PhD candidate or graduated with one of these degrees in 2015 or later, you are eligible to enter the competition!

Five papers (one each from Africa, Asia & Australasia, Europe, North America and South America) will be selected for presentation in the Young Statisticians’ Showcase session at the IBC. The winners will each receive a stipend of $3,000 (USD) to cover their expenses in attending the IBC in 2018.

For additional details on the application and selection processes, please click here.

To download and print the Young Statisticians Competition poster, please click here.

Call for Papers

The International Biometric Society (IBS) invites abstracts for contributed oral and poster presentations for the 2018 International Biometric Conference (IBC2018) being held in Barcelona, Spain from 19 January 2018Abstracts may be submitted online from 1 October 2017 to 10 January 2018. Submissions from a broad range of topics and perspectives are encouraged. See this announcement for a complete list of methodological topics and application areas that will be covered during the conference.

All abstracts must be submitted online. Each abstract submission must include the title, list of authors and author affiliations and abstract, with each submission having a maximum length of 2,500 characters. The number of oral abstract submissions with the same presenting author is limited to one. All submitted abstracts will be reviewed by the IBC2018 International Program Committee, and presenting authors will be informed of the status of their abstract (oral presentation, poster presentation, rejection) by 12 March 2018.

All abstracts must be submitted no later than 19 January 2018. For complete details, please click here.

Registration and Early Bird Deadlines

Registration is expected to open late December 2017 to early January 2018.  You will need your official IBS membership ID number to login to the IBS website to register online and receive IBS members-only pricing. Please visit our website for more information.  We will offer early-bird registration rates, so be sure to mark your calendars and do not miss the special pricing!

IBS Travel Grant Report

I, FOLORUNSO Serifat Adedamola from Department of Statistics, Faculty of Science, University of Ibadan, Nigeria.  A recipient of IBS Travel Awards Grant of 2000 USD (700,000 Nigeria Naira) for participation in The International Biometric Society Australasian Region (IBS-AR) conference “Biometrics by the Border” which was  held at the Mantra Hotel in Kingscliff, New South Wales, Australia 26th November – 30th November, 2017.

I applied for Visitor Visa on September 24, 2017 and the visa was granted on November 22, 2017, so my flight was booked on November 23, 2017. I travelled from Murtala Muhammed International Airport Lagos, Nigeria on Friday 24 November, 2017 to O.R.Tambo Johannesburg, then to Perth and finally to Brisbane Queensland in Australia on Monday 27, November, 2017.

There are excellent all female invited speakers which includes Elisabetta Carfagna from University of Bologna, Italy, Di Cook from Monash University, Australia, Racheal Fewster from University of Auckland, New Zealand, Sonja Greven from LM University Munich, Germany, Louise Ryan from University of Technology Sydney, Australia and Jean Yang from University of Sydney, Australia. The conference was opened by the current president of The International Biometric Society from Washington, USA, Elisabeth Thompson and closed by the in-coming president Louise Ryan from University of Technology Sydney, Australia. There are total of 120 delegates across the globe that participated in the conference.

The conference scientific session started on Monday and it was in parallel session with modeling session and sampling, complex surveys & mixed models session on Monday morning. General/Education session and Genetics Session for mid-day. Computing/Big data session and Environmental Statistics for Afternoon. Then Poster session followed at 16;30 which was sponsored by Grains Research and development Corporation. The same applied throughout the conference days with session like Missing data, high dimensional data, medical, modeling, fisheries, model selection, agriculture/horticulture, Biometrics, design, Multivariate, Bayessian, Biostatistics, genetics, and theory and methods.

I participated in all the activities of the conference from poster session on Monday; also I participated on other social activities such as the Excursion to Byron Bay Tourist Centre, young statistician dinner, and conference dinner.  I made a poster presentation on Thursday which was chair by the society in – coming president Louise Ryan and it was a great achievement for me as a young Biometrician.

The LOC including Ross Darnell, Alison Kelly, Clayton Forknall among others were so accommodating by giving me additional support in terms of accommodation and conference fee waiver, thanks to the executives for this offer. The IBS-AR was a good representative of the society in terms of hospitality and mentorship. I enjoyed my stay while in Australia and had opportunity to visit two Universities in Brisbane, University of Queensland and Queensland University of Technology. I am so delighted that I had a rewarding experience and good contacts while in the conference. I came back to my country on Tuesday night December 05, 2017.

Thanks to International Biometric Society for promoting the development and application of statistical and mathematical theory and methods in the biosciences, including agriculture, biomedical science and public health, ecology, environmental sciences, forestry, and allied disciplines. Also for sponsoring PhD student from developing countries in this non-IBC year, this has widened my scope as a research student and as a Biometrician.

I looked forward to IBC travel grant award as I will be willing to participate in IBC2018.

Thanks so much.

But I’m a Data Scientist Too, Aren’t I?

Louise Ryan, IBS Incoming President

Not long before being elected to serve IBS President, I received an invitation to give the Presidential Invited Address at the Spring 2017 meetings of the Eastern North American Region (ENAR). I was particularly excited because my former PhD student, Scarlett Bellamy now a full professor at Drexel University, was the incoming ENAR President and hence the one issuing the invitation! Also, during the many years that I had lived in the USA, ENAR had always been my favourite meeting, so I was excited by the chance to reconnect with many old friends and colleagues. Because I wanted to do a great job, I started my preparations well ahead of time (unusual for me!). After several practise talks at University of Technology Sydney where I work, I gave a version of my presentation in Melbourne at a branch meeting of the Statistical Society of Australia. While I wanted my presentation to have a sound methodological basis, I also wanted to be a bit provocative and talk about some of the larger issues facing us as a profession. Data science seemed like a natural focus. I started out a little tongue in cheek talking about how annoyed and frustrated I feel when people (especially bosses and funders!) get excited about “new” fields such as machine learning, analytics, data science, etc, ignoring that us statisticians have been doing many of the exact same things for years! I injected a little humour as well, quoting some funny definitions I found online, for example, that “a data scientist is a statistician who lives in San Francisco”. But then I got a little more serious, arguing that data science is more than just statistics dressed up with a new name and that it expands our more classical perspective to include a lot of modern and sophisticated computer science. An excellent article by David Donoho entitled “Fifty years of data science”1 talks about all this in a very interesting and provocative way, suggesting that for our profession to continue to thrive, we really do need to expand our perspective to embrace the increasing importance of computer science. I quoted two of my statistical heroes, John Tukey2 and Marvin Zelen, who long ago recognized the central importance of data and computing for the statistical profession.

After this general introduction, my talk dived into a few more detailed topics, drawing on some of my recent projects where statistics as usual did not quite work, due to the size and complexity of the data involved. Problems associated with the analysis of massive datasets can range from the data being too large to even fit on your machine, to the frequently encountered scenario where algorithms won’t run or take ages to converge, if they converge at all. While one solution is to upgrade your machine or access a high performance cluster, this isn’t always practical (we all know what bosses and IT departments tend to say when we ask for a new computer!). In my talk, I tried to argue that there are actually quite a lot of fairly simple and practical solutions that draw on sound statistical principles. Furthermore, seeking these kinds of solutions represents a great opportunity for us to contribute in a unique way to data science. The first thing I talked about was the so-called “Divide and Recombine” strategy that has been popularised through technologies like Hadoop®3. The basic idea is really simple. For example, when your data are too huge to run a standard analysis, a valid strategy will be to divide the data into subsets, perform an analysis on each subset and then average the results. The subsets can sit on the hard disk of your own machine or if really large, be distributed out on a cluster. Hadoop® and other similar technologies provide an infrastructure to manage the distributed data and do the various intermediate and final data analysis steps. While simple in principle, the learning curve to actually use such technologies can be steep, especially for us old-school statisticians who don’t have a lot of training in computer science. Luckily, there are a variety of efforts underway to make it a bit easier. I am particularly impressed by the R package DeltaRho4 developed by William Cleveland and various of his students and colleagues. If you haven’t come across it, take a look at the website where you’ll find some interesting tutorials and references to papers using the Divide and Recombine paradigm.

Learning about Hadoop and related concepts a couple of years ago got me thinking about whether there might be other simple strategies that draw on statistical ideas to ease the analysis of very large datasets. For example, we’ve been exploring whether concepts of sufficiency can be used to do sensible data reduction. In a paper published recently in the American Statistician5, we showed how to fit a Poisson random effects model to clustered count data, utilizing the concept of sufficiency to by-step the need to by create a large flat file of the kind needed by glmer() in R or Proc Genmod in SAS. Our results had an appealing side benefit in terms of privacy protection, since custodians of sensitive information need only provide analysts with a few key summary statistics rather than unit record data.

One of my favourite recent projects has involved a collaboration with the Blood Service of the Australian Red Cross. The goal here was to use their donor database to identify individual risk factors associated with vaso-vagal reactions (fainting) after blood donation, as well as to explore variations in these rates among the different donor centres. We wanted to fit a logistic regression model with random effects for donation centre, but the data were too large for the model to run. In a paper about to appear in Statistics in Medicine6, we explored the use of a case-control sampling strategy. With the addition of a simple adjustment to account for the sampling, we were able to estimate not only covariate effects of interest but also centre specific effects with surprising accuracy. Our approach had the appealing feature that it could utilize well known functions such as glm() and gamm(), with the simple addition of an offset term.

Currently we are exploring some other simple strategies such as coarsening to reduce data volume, without sacrificing too much in the way of statistical precision. Such strategies have also been used by computer scientists, but as statisticians, we have the tools available to figure out how to appropriately analyse the coarsened data. One of my current PhD students, Mr Hon Hwang, is exploring an approach that uses an EM algorithm and the results are looking quite promising and exciting.

I wrapped up my presentation by reiterating that there is indeed more to this whole data science world than statistics wrapped up with a new fancy name. There is a lot of impressive computer science there as well. We can sit back and moan about it, or we can rise to the challenge and expand our knowledge and capabilities accordingly. Many years ago, we used to think about a good biometrician needing to have a blend of mathematical/statistical and domain knowledge. In my talk, I argued that the mix has now expanded to include computer science as well (see diagram). We certainly should be making sure that our students are equipped with the right mix of skills to operate in this exciting new world.

Since giving my presentation at ENAR, I have had several other opportunities this year to talk about the ideas, including at the Eastern Mediterranean Region in May, at the Korean Region in June and again just recently at the Australasian Region. In all cases, the talk has been well received and has generated lots of constructive discussion. It is certainly clear that there are some great opportunities for our profession! Our hope really lies with the next generation who increasingly seem to be embracing modern computing as part of their arsenal of tools.


1   A version of David’s paper will be appearing soon in an issue of Computational and Graphical Statistics.

2 David also talked a lot about John Tukey in his article and in fact, David’s paper is a write-up of a presentation he gave a couple of years ago at a symposium to honour John.



5 Lee, Brown and Ryan (2017). Sufficiency Revisited: Rethinking Statistical Algorithms in the Big Data Era. The American Statistician 71: 202-208

6 Wright, Ryan and Pham (2017). A novel case-control subsampling approach for rapid model exploration of large clustered binary data. To appear in Stats. in Med.

Statistics and Food Security

How data analysis can help us feed the world.

By Daria Steigman

There will soon be 9.8 billion people.

According to a United Nations report released in June, 9.8 billion is the estimated world population in the year 2050. (It’s “only” 7.6 billion today.) And 2050 is not that far away.

The sharp population increase is frightening enough on its own, and even more so when you think about the challenge of sustaining 9.8 billion people with our limited natural resources. At the most basic level, we need a lot more food.

The good news is we’ve done this before. In the mid-20th century, the agricultural community put its best minds to work and underwent a green revolution. The exponential increase in food production at the time was credited to better fertilizers, new irrigation techniques, the use of pesticides, and other advances in farming.

This time, however, there are sobering constraints. There is less arable land thanks to urbanization and the tendency of people and governments to put up homes and factories on desirable (often agriculturally fertile) lands. There’s also less water. Moreover, we’re putting plants into biofuels, further dwindling our existing food resources. We’re also increasingly concerned about the environment (e.g., we don’t like pesticides) which, coupled with the impacts of climate change, mean we are less tolerant of and less interested in improving crop yields at the long-term expense of our water, our soil, and our air.

We’re in a quandary.

To get some answers, Biometric Bulletin talked with Kaye Basford, professor of biometry at the University of Queensland and a former president of the International Biometric Society. Basford was awarded a fellowship in 2006 with the Australian Academy of Technological Sciences and Engineering for having developed statistical methods that are widely used today to improve the efficiency and effectiveness of crop improvement programs.

“While we can increase food production, it is much harder to have the big expansion that we did during the green revolution,” says Basford. “That said, there are advances we can make to breed better crops, such as those that are less prone to disease, more drought-resistant, require less water, or are easier to harvest. We can also develop more nutrient-rich crops.”

And, yes, statistics has a role to play.

Statistics and Food Security

Basford highlighted various ways that statistics contribute to making agricultural advances and improving food security:

  • Designing for data generation—Basford noted, as an example, that multi-environment trials across locations, years, and management systems can be appropriately designed to generate terrific data sets for improving plant varieties.
  • Big data use—While “big data” has been a part of the food supply chain for a while, the sheer volume of information — combined with velocity, variety, volatility, validity, and veracity — has given biometricians far more breadth of data to use for real-time decision making.
  • Data modeling—Basford explained that statistical analyses that combine data from field experiments with data from high-throughput molecular marker technologies and from pedigree information can better determine ideal plant types. She added that simulation modeling (for example, of the genetic impact of repeated breeding cycles) can be used to identify the best options for plant improvement. Employing a variety of data sources and modeling approaches enable multidisciplinary teams to better understand a particular agricultural system.
  • Monitoring and evaluation—Basford noted that statistics is increasingly being adopted to assess the impact of agricultural activities, to track progress in yields, and to track progress in achieving sustainability goals (i.e., economic, social, and environmental impacts).

To illustrate the role of statistics in plant yields, Basford pointed to research that took the concept of a marker-trait association study and expanded on it to make the results interpretable under particular environmental and management conditions. To achieve this, Basford and her colleagues created a wheat phenome atlas.

“A phenome atlas is a collection of diagrammatic representations of chromosome regions that affect trait inheritance across the genome,” she said. “The wheat phenome atlas was constructed for a genome-wide association study of 20 economically important traits from the first 25 years of an international wheat breeding program.”

The result: researchers were able to identify genetic markers in the wheat for a wide range of factors, including heartiness of the crop and susceptibility to disease. It’s the kind of research that can—and does—lead to better yields.

Orange-fleshed Sweet Potatoes and Golden Rice

As much as the next food revolution must be about creating yields that can withstand current environmental conditions, it also has to be about how we can feed people will less. Because, as Basford made clear, we aren’t going to double food production again anytime soon.

One area that agricultural statisticians and scientists have been exploring is biofortified foods. These are plants bred to survive in specific conditions and fortified with key nutrients that people need but don’t have. For example, what if sailors in the 17th century had food stocks fortified with vitamin C to prevent scurvy?

The two best-known examples of biofortified foods are orange-fleshed sweet potatoes and golden rice. Three plant scientists shared the 2016 World Food Prize for their work in developing “disease-resistant, drought-tolerant, high yielding varieties of orange-fleshed sweet potatoes that can flourish in the variable soils and climatic conditions found in Sub-Saharan Africa.” The availability of these sweet potatoes will help to counter the effects of Vitamin A deficiency, including blindness, immune system disorders, and premature deaths of children and pregnant women.

Golden rice is another Vitamin-A biofortified food. The Golden Rice Project describes the plant as “obtained by genetic modification of the rice plant to produce and accumulate provitamin A in the grain.”

While there has been a lot of discussion about whether we should be genetically modifying our crops (a conversation for a different article), the reality is that for as long as people have grown food we’ve strived for better, stronger, more efficient yields. And today, there’s no doubt that we have a monumental task ahead if we’re going to coexist with Planet Earth and feed ourselves in the generations to come. And, however we achieve this, statistics will surely have played a key role.


March 2018 Issue Highlights

The March issue includes diverse articles from a range of applications and methodological areas.  The Biometric Methodology section features “Computation of ancestry scores with mixed families and unrelated individuals,” by Yi-Hui Zhou, J. S. Marron, and Fred A. Wright; “Estimating the size of an open population using sparse capture-recapture data,” by Richard Huggins, Jakub Stoklosa, Cameron Roach, and Paul Yip; “Covariate selection with group lasso and doubly robust estimation of causal effects,” by Brandon Koch, David M. Vock, and Julian Wolfson; “A Bayesian screening approach for hepatocellular carcinoma using multiple longitudinal biomarker,” by Nabihah Tayob, Francesco Stingo, Kim-Anh Do, Anna S. F. Lok, and Ziding Feng; and “FPCA-based method to select optimal sampling schedules that capture between-subject variability in longitudinal studies,” by Meihua Wu, Ana Diez-Roux, Trivellore E. Ragunathan, and Brisa N. Sanchez.

In Biometric Practice, articles include “Evaluating center performance in the competing risks setting: Application to outcomes of wait-listed end-stage renal disease patients.” by Sai H. Dharmarajan, Douglas E. Schaubel, and Rajiv Saran; “Estimating the probability of clonal relatedness of pairs of tumors in cancer patients,” by Audrey Mauguen, Venkatraman E. Seshan, Irina Ostrovnaya, and Colin B. Begg; “N mixture models vs Poisson regression,” by Richard J. Barker, Matthew R. Schofield, William A. Link, and John R, Sauer; and “Spatial Bayesian latent factor regression modeling of coordinate-based meta-analysis data,” by Silvia Montagna, Tor Wager, Lisa Feldman Barrett, Timothy D. Johnson, and Thomas E. Nichols.

As always, lists of papers to appear can be found at the Biometrics website.  Papers to appear in future issues may also be found under the “Early View” link at the Wiley Online Library website, which may be accessed by IBS members by visiting, selecting “Biometrics” from the drop-down menu at the “Publications” link at the top of the page, and accessing the “Click here” link.

Biometrics Showcase Session

A Biometrics Showcase Session will be held at the International Biometric Conference (IBC) in Barcelona, Spain, 8 – 13 July 2018.  As is convention, the session will feature the winners of the Best Paper Published in Biometrics by an IBS Member Awards for 2016 and 2017.  A selection committee of current and former Co-Editors will nominate the recipients of these awards in early 2018.

Journal of Agricultural, Biological, and Environmental Statistics (JABES) Editor Report

The new Twitter account @JabesEditor is helping to raise the profile of papers published in JABES. If you want to receive tweets about papers, or other issues related to JABES, please follow us.

The Special Issue on Animal Movement Modeling appeared in the September issue. It comprises the following articles by authors at the cutting edge of model development in response to technological changes:

“Guest Editor’s Introduction to the Special Issue on Animal Movement Modeling” by Mevin B. Hooten, Ruth King and Roland Langrock; “Multi-scale Modeling of Animal Movement and General Behavior Data Using Hidden Markov Models with Hierarchical Structures” by Vianey Leos-Barajas, Eric J. Gangloff, Timo Adam, Roland Langrock, Floris M. van Beest, Jacob Nabe-Nielsen and Juan M. Morales; “Incorporating Telemetry Error into Hidden Markov Models of Animal Movement Using Multiple Imputation” by Brett T. McClintock; “Selecting the Number of States in Hidden Markov Models: Pragmatic Solutions Illustrated Using Animal Movement” by Jennifer Pohle, Roland Langrock, Floris M. van Beest and Niels Martin Schmidt; “Hierarchical Nonlinear Spatio-temporal Agent-Based Models for Collective Animal Movement” by Patrick L. McDermott, Christopher K. Wikle and Joshua Millspaugh; “Modeling Collective Animal Movement Through Interactions in Behavioral States” by James C. Russell, Ephraim M. Hanks, Andreas P. Modlmeier and David P. Hughes; “Imputation Approaches for Animal Movement Modeling” by Henry Scharf, Mevin B. Hooten and Devin S. Johnson; “Reflected Stochastic Differential Equation Models for Constrained Animal Movement” by Ephraim M. Hanks, Devin S. Johnson and Mevin B. Hooten; “Bayesian Inference for Multistate ‘Step and Turn’ Animal Movement in Continuous Time” by A. Parton and P. G. Blackwell; and “A General Approach to Model Movement in (Highly) Fragmented Patch Networks” by Juan Manuel Morales, Agustina di Virgilio, María del Mar Delgado and Otso Ovaskainen.

If you have a suggestion for a special issue, I would be pleased to hear from you. We are also keen to publish papers that summarize the state of methodological development in subject areas for which technological advances are generating a demand for new statistical approaches. If such papers also speculate on likely future developments, so much the better. If you feel that you could offer such a paper, or can suggest a topic together with possible authors, please let me know.

For more information on upcoming issues, the editorial board, and the aim and scope of the journal, please visit our website We also accept submissions of books to review in the upcoming issues of JABES; to submit a book for review, please see the above website (click on “Editorial Board”) or contact Ken Newman (

Steve Buckland, Editor in Chief

Software Corner

Superheat: creating beautiful and extendable heatmaps for visualizing complex data in R

Rebecca Barter and Bin Yu

An extended version of this article by authors Barter and Yu can be found on ArXiv ( and has been tentatively accepted by the Journal of Computational and Graphical Statistics (JCGS).

The heatmap as a general visualization tool

Extracting useful information from the massive datasets of today’s era is an ongoing challenge as traditional data visualization tools such as scatterplots, boxplots, and histograms rarely extend well in high-dimensional settings. An existing visualization technique that is particularly well suited to visualizing large datasets is the heatmap. Although heatmaps are extremely popular in fields such as bioinformatics, they remain a severely underemployed visualization tool in modern data analysis.

A heatmap can be used to visualize a data matrix by representing each matrix entry by a color corresponding to its magnitude, enabling the user to visually process large datasets with thousands of rows and/or columns.

Extending the heatmap using superheat

Inspired by a desire to visualize multiple types of data simultaneously (such as a design matrix with model residuals), we developed a new R package called superheat that provides an extremely flexible and customizable platform for visualizing complex datasets using the heatmap as its foundation.

Beyond the traditional heatmap, superheat offers a number of novel features such as smoothing of the heatmap within clusters to facilitate extremely large matrices with thousands of dimensions; overlaying the heatmap with text or numbers to increase the clarity of the data provided or to aid annotation; and most notably, the addition of scatterplots, barplots, boxplots or line plots adjacent to the rows and columns of the heatmap, allowing the user to add an entirely new layer of information such as a re.

Superheat examples

Some of these features are shown in the example below which shows the rate of organ donation per year for 20 countries. This data was collected from the WHO-ONT Global Observatory on Donation and Transplantation. In the center of the figure is the heatmap which shows how many organs were donated per 100,000 individuals yearly for each country (country name is provided in the labels on the left-hand side of the plot). The country labels are colored by geographic region.

Above the heatmap, a line plot (with scatter points) shows the total number of transplants per year aggregated over all countries (not just the 20 presented), and the barplot on the right-hand side displays the human development index ranking for each country (a lower ranking is better). This ability to add additional information, along with the ease of customization are what makes superheat such a powerful visualization tool.

The superheatmap shows very clearly a number of things including the fact that Spain is clearly a leader in organ donations, Croatia has drastically improved its rate of organ donation over the past decade and that countries with higher rates of organ donation tend to have better (lower) HDI rankings.

Another example is shown below displays the cosine similarity matrix of word vectors obtained using the word2vec algorithm applied to the Google News dataset. Words that appear in similar contexts (such as “deal” and “talks” as well as “president” and “leader”) are shown in this example to have very high cosine similarity. This example displays the default viridis color scheme.


Along with the package is an extremely detailed vignette ( which guides the user through the wide variety of features that superheat provides.

While superheat is available on CRAN, the most recent version can be found on github. Usage of superheat itself is simple as the example below shows. The following code installs and loads superheat and then uses superheat to produce the organ donation figure above. The pre-processing code can be found here:

# install devtools if you don’t have it already
# install the development version of superheat
# use superheat
# set heatmap color map
heat.pal = brewer.pal(5, “BuPu”),
# grid line colors
grid.vline.col = “white”,
# right plot: HDI
yr = hdi.match.2014$rank,
yr.plot.type = “bar”, = «Human Development Ranking»,
yr.obs.col = region.col,
# top plot: donations by year
yt =,
yt.plot.type = “scatterline”, = «Total number of transplants per year»,
# left labels
left.label.col = adjustcolor(region.col, alpha.f = 0.3),
# bottom labels
bottom.label.col = “white”,
bottom.label.text.angle = 90,
bottom.label.text.alignment = “right”)

STRengthening Analytical Thinking for Observational Studies (STRATOS): Introducing the Missing Data topic group (TG1)

James Carpenter (1) and Katherine J Lee (2) on behalf of STRATOS TG1.

(1)  London School of Hygiene & Tropical Medicine and MRC Clinical Trials Unit at UCL. Email:

(2)  Murdoch Children’s Research Institute, Melbourne and Department of Paediatrics, University of Melbourne. Email:

In the last issue of the Bulletin, the aims and recent activities of  the STRATOS initiative were described. This initiative is divided into a number of topic groups, which provide a focus for specific research methodologies. Here, we introduce the Missing Data topic group.

Missing data are ubiquitous in medical and social research, and there is a long history of methods for coping with the difficulties they raise. As the Figure illustrates, early work (dating back at least to 1926) was largely computational, addressing the question of how to perform the analysis when data are missing. Later, the focus switched to establishing a valid inferential framework when data are missing, giving rise to an array of methods for handling missing data including multiple imputation, inverse probability weighting and Bayesian methods. This research has also linked with work in non-parametric statistics and gave rise to doubly robust estimation (which seeks to combine the best of imputation and inverse probability weighting).  How best to handle missing data remains an active area of research, which has recently been brought together in the Handbook of Missing Data Methodology (Molenberghs et al, 2015).

Most researchers, whether or not they have a formal statistical training, are aware of the challenges missing data raise, and at least some of the approaches available. In particular, multiple imputation has become increasingly popular for handling missing data, fuelled by the development flexible software that is now available in all the leading statistical software packages.

However, in our experience, many researchers remain unclear about the practical value of more sophisticated approaches like multiple imputation compared with restricting the analysis to those with no missing data. Often, they may also be unsure about how methods like multiple imputation, full information maximum likelihood, the Expectation-Maximization algorithm and inverse probability weighting relate to each other, and the practical implications of this in their specific setting.

The Missing Data topic group is working to address these issues through a series of linked papers. Following the aims of the initiative (Sauerbrei et al,2014( we seek to address three audiences: those engaged in quantitative research, but without a formal statistical training (level 1); those with a statistics training to masters level (level 2), and those interested or active in missing data research (level 3).

The topic group consists of researchers known internationally for their work across the broad spectrum of theoretical and practical methodology for missing data: Rod Little (Michigan, USA),  Andrea Rotnitzky (Harvard, USA), Joe Hogan (Brown, USA) Els Goetghebeur (Gent, Belgium), Ian White (UCL, London), Kate Tilling (Bristol, UK) and Melanie Bell (Arizona, USA). James Carpenter (London, UK) and Katherine Lee (Melbourne, Australia) chair the group.

The topic group played an active role in the 2016 STRATOS workshop at the mathematical sciences research centre in Banff, Canada, where James gave an overview lecture  ‘Handling missing data in observational studies: challenges for teaching and research’, which is publicly available at Building on this, currently the Topic Group has three papers close to submission, which will be highlighted on the STRATOS website when they are available (

The first paper, led by Rod Little, compares three popular methods for handling missing data in a social science setting: complete cases, weighting and multiple imputation. This paper is aimed at researchers with relatively little formal statistical training, and uses a simple example from the UK’s Youth Cohort Study, a publically available dataset, to build intuition for how biases may be caused by missing data, and the pros and cons of these approaches. It includes practical guidance on which methods are preferable in which situations.

The second paper, led by Katherine Lee, targeted at those using observational data for medical research, is likewise aimed at researchers with relatively little formal statistical training. Building on related work in clinical trials (which is primarily concerned with missing outcome data), this paper aims to provide and illustrate a practical framework for the analysis of partially observed data and subsequent reporting. Again this paper will include a worked example from the Youth Cohort Study along with example code.

The third paper addresses statisticians, and discusses both theoretically and with examples, the utility of key approaches to the analysis of partially observed data, in particular: full Bayesian analysis, multiple imputation, inverse probability weighting, doubly robust estimation, direct maximum likelihood and the EM algorithm. This paper will include the statistical code for each of the approaches and will be based on an example from a publically available dataset so that readers will be able to re-create the analyses presented. The aim in this paper is to providing the necessary tools for researchers wishing to apply these approaches in practice.

As missing data, and in particular the most appropriate way to handle the issues it raises, varies depending on the setting and the statistical models required for analysis, TG1 is also linked to other groups in the STRATOS initiative, particularly the Initial Data Analysis (STRATOS Topic Group 3), Measurement Error Misclassification (STRATOS Topic Group 4) and Causal Inference (STRATOS Topic Group 7).

The Missing Data Topic Group is keen to stimulate interactions with other researchers; James and Katherine welcome any comments and suggestions on its work, future issues that would be useful to consider and illustrative datasets.

References in text

Sauerbrei, W., Abrahamowicz, M., Altman, D. G., le Cessie, S. and Carpenter J R on behalf of the STRATOS initiative (2014). STRengthening Analytical Thinking for Observational Studies: the STRATOS initiative. Statist Med, 33, 5413-5432.

Molenberghs, G., Fitzmaurice, G., Kenward, M. G., Tsiatis A. and Verbeke, G (2015). Handbook of Missing Data Methodology. New York: CRC press

References for Figure

Dempster, A. P., Laird, N. M. and Rubin, D. B. (1977) Maximum likelihood from incomplete data via the EM algorithm (with discussion). Journal of the Royal Statistical Society, Series B, 39, 1-38.

Healy, M. J. R. and Westmacott, M. (1956) Missing values in experiments analyzed on automatic computers. Applied Statistics, 5, 203-206.

Heckman, J. J. (1976) THe common structure of statistical models of truncation, sample selection and limited dependent variables and a simple estimator for such models. Annals of Economic and Social Measurement, 5, 475-492

Horvitz, D. G. and Thompson, D. J. (1952) A generalisation of sampling without replacement from a finite universe. Journal of the American Statistical Association, 47, 663-685.

Laing, K-Y and Zeger, S. L. (1986) Longitudinal data analysis using generalized linear models. Biometrika, 73, 13-22.

Louis, T. (1982) Finding the observed information matrix when using the EM algorithm. Journal of the Royal Statistical Society Series B, 44, 226-233

McKendrick, A. G. (1926) Applications of mathematics to medical problems. Proceedings of the Edinburgh Mathematical Society, 44, 98-130.

Orchard, T. and Woodbury, M. (1972) A missing information principle: theory and applications. Proceedings of the Sizth Berkely Symposium on Mathematics, Statistics and Probability, editors Le Cam, L. M., Neyman J. and Scott, E. L, 1, 697-715

Robins, J. M. and Gill, R. (1997) Non-response models for the analysis of non-monotone ignorable missing data. Statistics in Medicine, 16, 39-56.

Robins, J. M., Rotnitzky, A. and Zhao, L. P. (1995) Analysis of semiparametric regression models for repeated outcomes in the presence of missing data. Journal of the American Statistical Association, 90, 106-121.

Rubin, D. B. (1976) Inference with missing data. Biometrika, 63, 581-592.

Rubin, D. B. (1987) Multiple imputation for non-response in surveys. New York: Wiley

Scharfstein, D. O., Rotnitzky, A., and Robins, J. M. (1999) Adjusting for non-ignorable drop-out using semiparametric nonresponse models.

Schafer, J. L. (1997) Analysis of incomplete multivariate data, London: Chapman and Hall

van Buuren, S. and Boshuizen, H. C. and Knook, D. L. (1999) Multiple imputation of missing blood pressure covariates in survival analysis. Statistics in Medicine, 18, 681-694.


Solution to the Mathematical Riddle of Vol 34 3rd issue

Let’s see if you can solve this riddle – What is the larger number you can create by moving only two matches? The digits in this number should be similar in size. You are not allowed to change the place of digits, just to move two matches.

* Please ignore the Hebrew! IBS does not reserve the rights for this riddle.

The solution to the last issue’s mathematical riddle was:

The five individuals who answered correctly were:

  1. Michael Mcisaac, Department of Public Health Sciences, Queen’s University, Kingston, Ontario, Canada
  2. Almuth Marxs, Nurnberg, Germany
  3. Michal Sharabi, Volcani Center, Agriculture Research Organization, Israel
  4. Nisim Mery, Israeli Center for Disease Control, Israel
  5. Tehila Ben-Tal, Gertner Institute, Israel

Please email interesting riddles to be published in the future issue to

Region News

Two New Regions Join the IBS Family of Regions

By Peter Doherty

During the Executive Board’s final meeting of the year on 24 October, two requests for Region status were approved. Each Region was reported to have completed the minimum requirements.

The formation of a new North African Region, to be known as NAR and representing Morocco, Algeria and Tunisia, was approved by the Board. Hamid El Maroufy, newly elected President of the Region, and Ziad Taib, President of the Nordic/Baltic Region, were instrumental in the recruitment for and formation of this Region.

Meanwhile, the Tanzanian Region, to be known as TZR, has met all of the requirements and has also been approved by the Board. Newly elected President Innocent Mboya and Dr. Jim Todd were successful in recruiting many new Region members to the IBS.

As stated in the IBS Bylaws, the Society may, from time to time, “establish an official Region in a geographically defined area in accordance with the procedures set forth in the Policies and Procedures of the Society.” The IBS is pleased to welcome these two new Regions to the growing list of IBS Regions. Congratulations to both of our new Regions, and best wishes for future success!

Australasian Region (AR)

Andrew Zhou (President IBS-China) and Samuel Mueller (President IBS-AR) met during the 2017 Joint Statistics Meeting in Baltimore to discuss new ways to strengthen the two regions. First steps include that a small contingent of IBS-China members will attend our November “Biometrics by the Border” conference and it is planned that a contingent from the Australasian region will visit the 2018 conference of the China region.

Zheyu Wang (Johns Hopkins), Andrew Zhou (President of IBS-China) and Samuel Mueller (President of IBS-AR).

Joint International Society for Clinical Biostatistics and Australian Statistical Conference, Melbourne, August 2018

On behalf of the International Society for Clinical Biostatistics (ISCB) and the Statistical Society of Australia (SSA), the organizing committee invites you to the 2018 Joint International Society for Clinical Biostatistics and Australian Statistical Conference, which will take place in Melbourne from 26-30 August 2018. The aim of this conference is to bring together a wide range of statistical researchers and practitioners to facilitate the international exchange of theory, methods and applications.

This meeting, which will be the 24th Australian Statistical Conference (ASC) and the 39th conference of the ISCB, promises to be a fascinating event as Big Data becomes commonplace, precision medicine becomes palpable with rapidly advancing “omics” technologies, census-taking and social surveys face unprecedented pressures, environmental concerns place ecological research under increasing strain, and online social interaction becomes the norm. Have statistical methods kept pace with societal change?

The conference will include talks on a broad spectrum of topics including modern analytics, Bayesian methods, high dimensional data, causal inference, clinical trials, statistical methods for precision medicine and many more. The scientific program, developed by a committee led by John Carlin and Tim Brown, promises to be of interest to participants from universities, public institutions, government and industry alike.

Highlights of the program are:

  • Four eminent keynote speakers: Chris Holmes (University of Oxford), Susan Murphy (Harvard University), Louise Ryan (University of Technology Sydney) and Thomas Lumley (University of Auckland).
  • An exciting range of invited sessions, including sessions on environmental statistics, statistical genomics, robust Bayesian inference, precision medicine, cluster cross-over trials, time-to-event analysis and statistical learning methods for causal inference.
  • A stimulating mix of half-day and full-day pre-conference courses.
  • An Early Career Researcher’s Day, which will provide a unique opportunity for early career researchers to share experiences and to practise presentation skills in a less formal environment.

Registrations and Call for Abstracts open in mid-November. For more information, please visit our website at

Australasian Applied Statistics Conference, Rotorua, December 2018

Save the date!  Join us for Australasian Applied Statistics Conference and pre-conference workshops, being held from 3-7 December 2018 at Millennium Hotel, Rotorua, New Zealand.

Vanessa Cave

Belgian Region (RBe)

Summer School on Advanced Bayesian Methods

The Interuniversity Institute for Biostatistics and statistical Bioinformatics organized with the Belgian branch of the Biometric Society, the International Society for Bayesian Analysis and the International Society for Clinical Biostatistics for the first time a Summer School on Advanced Bayesian Methods. During one week, two courses were taught at the Catholic University of Leuven (Belgium) on specific topics in Bayesian methodology with special attention devoted to novel statistical methodology. In this first edition, a three day course was given by Dr. Alejandro Jara (Pontificia Universidad Católica, Chile) on nonparametric Bayesian methods and a two day course was given by Dr. Gary Rosner (Johns Hopkins University, US) on Bayesian clinical trials. These two courses attracted 33 participants with 16 nationalities.

Participants to the Bayesian Clinical Trials workshop with Dr. Gray Rosner (left) and Dr. Emmanuel Lesaffre (right)

The second edition will be  held in Leuven too  24 -28 September2018. The program includes a three day course on Bayesian Parametric and Nonparametric Methods for Missing Data and Causal Inference given by Dr. Michael Daniels (University Florida, US) and a two day course on Bayesian Methods in Health Economic Evaluation given by Dr. Gianluca Baio from University College London, UK. More information can be found at

25th Annual meeting of the Belgian Statistical Society.

The 25th Annual Meeting of the Belgian Statistical Society was held in the beautiful Irish College in Leuven during 18-20 October 2017. The Quetelet Society (Belgian Branch of the Biometric Society) organized two sessions, meeting with great success. During the first session, two winners of the Quetelet prize presented their work. The Quetelet prize is a possibility given to Institutional members of the Quetelet Society to award one or more students for the excellence of their master thesis in Statistics. Zoé Pieters (Hasselt University, prize winner 2015-2016) investigated the influence of maternal pre-gestational BMI on the newborn’s gene expression and birth weight using the ENVIRONAGE birth cohort. Then, the 2016-2017 winner from Hasselt University, Daniel Olusoji Oluwafemi, presented his work on the evaluation of surrogate endpoints in Human microbiome. Both are now PhD students! The 2016-2017 Ghent University laureate, Stijn Decubber, was unfortunately not able to present his work on Granger causality methods for climate change attribution. During the second session, the invited speaker Hélène Jacqmin-Gadda (Bordeaux University) gave a very interesting talk on quantile regression for incomplete longitudinal data with selection by death. She illustrated the procedure on the French cohort Paquid of elderly to estimate cognitive norms by age taking into account selection by death.

The two Quetelet prize winners, Daniel Olusoji Oluwafemi (left) and Zoé Pieters (right) with the president of the Quetelet Society (Sophie Vanbelle)

Central European Network (CEN)

CEN-ISBS Vienna 2017 – Joint Conference on Biometrics and Biopharmaceutical Statistics

After Munich 2008 and Zurich 2011, the third conference of the Central European Network (CEN) was held in conjunction with the International Society for Biopharmaceutical Statistics (ISBS) at the Medical University of Vienna from August 28 to September 1, 2017. CEN consists of the German Region (IBS-DR), the Austro-Swiss Region (IBS-ROeS) and the Polish Region (IBS-PR). The more than 700 attendees were warmly received by the local organizing committee, led by Georg Heinze. The scientific programme committee was formed by the collaborating regions and societies IBS-DR, IBS-ROeS, IBS-PR and ISBS and co-chaired by Martin Posch, Vienna (IBS-ROeS), Tim Friede, Göttingen (IBS-DR), Tomasz Burzykowsky, Hasselt (IBS-PR) and Frank Bretz, Basel (ISBS). The logistic challenges in planning and organizing this joint conference of different societies, each with its own tradition, were finally mastered by a great collaboration between the scientific and local organizing committees.

After pre-conference courses on ‘Joint modelling of longitudinal and time to event data’ (Dimitris Rizopoulos), ‘Bayesian pharmaceutical applications using SAS(R)’ (Fang Chen, Frank Liu), ‘Interval-censored time to event data’ (Din Chen, Tony Sun), and ‘Statistical evaluation of surrogate endpoints’ (Tomasz Burzykowsky), the conference started with a keynote lecture by John P. A. Ioannidis on ‘Conceptional and statistical issues on reproducibility’, recognized by Austrian mass media. Further keynote lectures were given by Ulrich Dirnagl (‘Statisticians to the rescue: a humble stroke researcher’s proposal how to improve the quality of preclinical biomedicine’), Allison Smith (‘Design tableau: an aid to specifying the linear mixed model for a comparative experiment’) and Stijn Vansteelandt (‘Inferring causal pathways from data: challenges and some solutions’).

The conference featured 19 further invited talks, given on invitation of the scientific programme committee, as well as 52 topic-contributed sessions and 22 contributed sessions. Particular highlights of the conference were four sessions dedicated to young statisticians, including two award sessions from the German and Austro-Swiss regions. We were also happy to provide travel support to our young academics and to researchers from lower-income countries. On the last two days, the European Medicines Agency joint our conference by co-organizing a symposium with 9 further sessions. Several meetings of the various working groups, as well as a meeting on the ‘Future of CEN – interregional activities and new members’, exploring the possibilities of extending our activities beyond current limits, provided opportunities for fruitful discussions. Two further courses on ‘Evaluating therapies in rare diseases’ and ‘Confirmatory adaptive designs’, and a poster (and wine) session complemented the scientific programme.

Talking about wine, also the social programme provided several highlights and reinforced possibilities for communication and networking among the 700 delegates. On Monday, some city walks were organized which featured the Ringstrasse and the famous ‘Palaisviertel’, a neighborhood in the inner city that boasts an unmatched wealth of private palaces. After the city tours, young researchers gathered at an informal dinner in the former General Hospital of Vienna, now partly owned by the university and by private businesses, a hot-spot of urban nightlife. Tuesday closed with a welcome reception at the magnificent Great Ball Room of Vienna’s City Hall. The traditional hiking excursion of the ROeS led through the Wienerwald (Vienna Woods), passing at the highest ‘summit’ of Vienna, and finally ending at a traditional wine tavern. This was also the venue of the conference dinner on Thursday, which, thanks to good weather, could take place outside in a beautiful garden, accompanied by a musical performance of an uprising Viennese singer-songwriter.

A selection of photographs and the full conference programme can be found at the conference website

Georg Heinze

ROeS Arthur-Linder-Award winners

The local organizers

Award winners of the German Region

Hiking in the Vienna Woods

Panel discussion on reproducibility – Keynote speakers Ulrich Dirnagl and

John P.A. Loannidis, University President Markus Müller, Stephen Senn

At the Welcome Reception (City Hall)

ISBS president Jie Chen at the Opening Session

Eastern Mediterranean Region (EMR)

EMR-IBS Conference 18-20 December 2018 Jerusalem, Israel

We are very glad to welcome you to the: Jerusalem Joint Statistical Event 2018.

The event will include a one-day symposium in honor of the 70th birthday of Professor Yoav Benjamini of Tel Aviv University, followed by the 10th Conference of the Eastern Mediterranean Region (EMR) of the IBS. The conference will be organized by Malka Gorfine (Tel-Aviv University, Israel) and David Zucker (The Hebrew University of Jerusalem, Israel). Visit the
website for further information.

Eastern North American Region (ENAR)

2018 ENAR Spring Meeting, Atlanta, GA, USA

The 2018 Spring Meeting of the IBS Eastern North American Region, in conjunction with the Institute of Mathematical Statistics (IMS) and sections of the American Statistical Association (ASA), will be held 25-28 March at the Hyatt Regency Atlanta on Peachtree St. Located in the heart of downtown Atlanta, this luxury hotel is close to many of the city’s attractions, including Centennial Olympic Park, the Center for Civil and Human Rights, and the World of Coca-Cola, and is within walking distance of outstanding restaurants and entertainment.

The scientific program will once again be phenomenal, with a wide variety of topics including analysis of tracking data from wearable devices, network meta analysis, machine learning methods for imaging data analysis, novel clinical trial designs, statistical modeling to address human rights issues, multi-omics and graphical models for precision medicine, and novel extensions of causal inference methods. The Presidential Invited Speaker will be Dr. Roderick Little, Richard D. Remington Distinguished Professor, Department of Statistics, University of Michigan. The title of his talk is “Statistics as Prediction” and will provide an “illuminating discussion of how prediction can serve as a unified motivation to all of statistical inference.” Dr. Little is well known for his methodological contributions to the analysis of data with missing values and model-based survey inference with applications in medicine, demography, economics, psychiatry, aging, and the environment. His book “Statistical Analysis with Missing Data,” coauthored with Donald Rubin, has been cited over 22,000 times. He is a recipient of the American Statistical Association’s Wilks Medal and a past COPSS Fisher Lecturer at the Joint Statistical Meetings. Dr. Little is a Fellow of the American Statistical Association and the American Academy of Arts and Sciences, and a member of the National Academy of Medicine.

A complete listing of the many invited sessions to be presented at the meeting can be found at In addition, the program will feature both full and half-day short courses: “Bayesian Adaptive Clinical Trials” (Scott Berry.); “Machine Learning for Biomarker Discovery” (Noah Simon); “An Introduction to Joint Modeling of Longitudinal and Survival Data with Applications in R” (Dmitri Rizopoulos); “Neuroimaging Analysis within R” (John Muschelli, Jean-Phillipe Fortin); “Causal Inference: Structural Nested Mean Models” (Judith Lok); “ Reproducible Research in R” (Keith Baggerly); and “Survival for Precision Medicine” (Lu Tian, Lihui Zhao).

Several tutorials will once again be offered, running concurrently with the scientific sessions. The topics range from micro-randomized trials for mobile health intervention to integrative analysis of high through-put multi-platform genomics data. Additionally, roundtables will allow an opportunity to interact with experts and peers in a less formal setting. The roundtables offer a variety of topics, both professional development and statistical, and are an opportunity to interact with some of the outstanding ENAR leaders!

ENAR 2018 will feature one pre-conference workshop, Fostering Diversity in Biostatistics, which will take place on Sunday afternoon. Sunday evening will feature the new member reception, opening mixer and poster session, during which the ENAR Regional Advisory Board poster competition will be held. The Council for Emerging and New Statisticians will hold a student mixer on Monday evening, and the Career Placement Center will take place throughout the meeting to offer assistance to those seeking employment.

Special thanks to those who are working hard to help plan the ENAR Spring Meeting. Program Chair Veera Baladandayuthapani( and Associate Chair Jeff Goldsmith ( led a tremendous effort in crafting the scientific program. This year’s Program Committee was very active and included representation from 13 sections of the American Statistical Society as well as the Institute of Mathematical Sciences. The Education Advisory Committee (Simone Gray, Ali Shojaie, Lorenzo Trippa, Wenyi Wang, and Zhenke Wu) has been instrumental in developing an exciting offering of short-courses, tutorials and roundtables, and the Local Arrangements Chair, Howard Change has identified interesting adventures in and around Atlanta to keep you busy during your meeting downtime.

2018 JSM Vancouver, British Colombia, Canada

The 2018 Joint Statistical Meetings will be held in Vancouver, British Colombia, Canada from 28 July – 2 August 2018, and ENAR is fortunate to have Brian Reich be our representative to the Program Committee. Questions should be directed to Brian at

2019 ENAR Spring Meeting, 24-27 Philadelphia, PA, USA

Stay tuned for information about the 2019 ENAR Spring Meeting in Philadelphia, PA!

ENAR Webinar Series! Details about upcoming ENAR webinars can be found at

Please contact Elena Polverejan ( if you have suggestions for webinar topics.

French Region (RF)

The French Region organized two meetings in 2017. The first one was in June in Avignon within the Conference of the French Statistical Society ( A morning session featured three distinguished speakers, Sylvia Richardson invited by the French Statistical Society (Statistical genomics: recent developments and future challenges), Andrew Titman (Testing the Markov assumption in general multistate models) and Pascal Monestiez (Modélisation statistique des données d’observation issues des sciences participatives).The second one was a two day conference jointly organized with the “Groupe de Recherche Statistique et Santé”  in Bordeaux in October 5-6. There were four invited speakers: Avner Bar-Hen (Sélection de variables avec données manquantes), Paul Blanche (Constrained nonparametric maximum likelihood estimates in the competing risks setting), Laurent Jacob (Representing genetic determinants in bacterial GWAS with compacted De Bruijn graphs) and Thomas Gerds (Risk re-classification boxplots); see photos of the first two.

The French Region plans to repeat these two meetings in 2018. A session will be organized within the conference of the French Statistical Society in Paris ( (May 28 – June 1). As for the joint conference with the GdR, it will be held in Nantes.

The French Region is a member of the Channel Network which includes four Regions, the French, British and Irish, the Belgian and the Netherlands. The Channel Network Conference ( was held in Hasselt-Diepenbeek in April 24-26, 2017. The next one in 2019 is planned to happen in Rothamstead, for a commemoration of Ronald Fisher.

Finally, with the aim of promoting biometric research in young researchers, it has been decided to organize a Conference for young researchers in Biometrics and to launch a thesis award named “Prix de thèse Daniel Schwartz” which will be given at this occasion.  The key-note speaker will be Geert Molenberghs. The conference is planned to be held on April 3, 2018 in Paris (at the CNAM).

More information can be found on our website

Paul Blanche,  Invited speaker at the joint conference SFB – GdR Statistique et Santé held in Bordeaux, October 5-6, 2017.

Avner Bar-Hen,  Invited speaker at the joint conference SFB – GdR Statistique et Santé held in Bordeaux, October 5-6, 2017.

German Region (DR)


A joint workshop was held on October 4-5 in Ulm, organized by: The Institute of Statistics of Ulm University and the IBS-DR working groups: “Bayes Methodik” (Bayes Methods) and “Pharmazeutische Forschung” (Pharmaceutical Research). The speakers were Beat Neuenschwander and Simon Wandel, Novartis Pharma AG, Basel. The workshop was aimed to provide an introduction to the application of Bayesian methods in clinical trials. Main topics were Bayesian meta-analysis, prior derivation, hierarchical models and probability of success.

More than 70 participants, who had a quite broad range of experience with Bayesian methods, took the opportunity to learn and to intensively discuss recent Bayesian methods for clinical trials in the pharmaceutical industry context.

(from left to right): Frank Fleischer (for the WG Pharmaceutical Research), Beat Neuenschwander and Simon Wandel (both Novartis Pharma AG), Jan Beyersmann (Institute of Statistics). Photographer: Reinhard Vonthein (WG Bayes Methods)

Japanese Region (JR)

The 2017 Japanese Joint Statistical Meeting

The Biometric Society of Japan (BSJ) was one of the six sponsoring organizations of the 2017 Japanese Joint Statistical Meeting that was held on 3-6 September at Nanzan University in Nagoya, Japan. The ASA’s statement on p-values: context, process, and purpose have much impact on the statistical society in Japan, and two invited sessions on this topic were organized: The first on “P-values in medical and agricultural research: beyond the p<0.05 paradigm”, organized by BSJ as the Biometric Symposium, included discussion on P-values from the perspective of epidemiology by four researchers: Prof. Tosiya Sato (Kyoto University), Prof. Nobuhiro Minaka (National Institute for Agro-Environmental Sciences), Prof. Shigeyuki Matsui (Nagoya University) and Prof. Takashi Sozu (Tokyo University of Science). The society also organized an invited session, in which the two winners of the Young Biostatisticians Award conferred by the society made their presentations on their research published in Japanese Journal of Biometrics, the official journal of the BSJ. Ms. Kasumi Iwamoto (The Chemo-Sero-Therapeutic Research Institute) talked about their findings in statistical methods for diagnostic medicine and Dr. Junji Moriya (Kyowa Hakko Kirin Co., Ltd.) talked about their new proposal concerning the principal stratification approach in causal inference and its application.

The Biometric Seminar

The Biometric Seminar entitled “Estimand in clinical trials and sensitivity analysis” will be held on December 15, 2017 at TKP GARDEN CITY PREMIUM Jimbocho, Tokyo, Japan. The seminar is being organized jointly by the BSJ and Clinical Biostatistics Course, Kyoto University. The seminar topic is the recent update of the ICH E9 (R1) guideline. Perspectives from each ICH region will be presented by Dr. Yuki Ando (Pharmaceuticals and Medical Device Agency), Dr. Thomas Permutt (US Food and Drug Administration) and Dr. Frank Bretz (Novartis).

Satoshi Hattori

Polish Region (GPol)


On the 10-14 September 2017, the 47th International Biometrical Colloquium was held in the Marymont Training and Lesure Centre in Kościelisko-Kiry, near Zakopane – the capital of Tatra Mountains. The Colloquium was organized by the Department of Experimental Design and Bioinformatics of Warsaw University of Life Sciences, supported by the Polish Biometric Society, the Faculty of Mathematics and Computer Science at Adam Mickiewicz University and the Department of Mathematical and Statistical Methods in Poznań. The chairman of the organizing committee was Assistant Professor Dariusz Gozdowski. About 54 biometricians from Poland, Czech Republic and Slovakia attended this conference (see photo). During the conference 30 oral presentations and 14 posters were presented.

Some of the presented papers are published in the 47th volume of the Colloquium Biometricum (Online), the official periodical of the Polish Biometric Society. The scientific atmosphere of the conference was enriched by hiking trips in the Tatra Mountains.

The next, 48th International Biometrical Colloquium will be organized in Poznań, by the Department of Mathematical and Statistical Methods at University of Life Sciences. I have the pleasure to inform you that during this Colloquium we will celebrate the 90th birthday of an outstanding Professor – Tadeusz Caliński.

We strongly encourage biometricians to participate in this event (the conference fee is about 210 Euro), and also to send scientific papers to publish in Biometrical Colloquium (Online).

Zofia Hanusz

Western North American Region (WNAR)

2018 WNAR/IMS meeting

The 2018 WNAR/IMS meeting will be in Edmonton, Canada from 24-27 June at the University of Alberta.  The campus is located on the southern bank of the North Saskatchewan River.  As one of the largest cities in Canada, Edmonton is a cultural center, with many arts and culture events anchored in the downtown Arts District, accessible from campus by the city light rail system.  Both the Edmonton Jazz Festival and Freewill Shakespeare Festival are scheduled to occur in the city during the WNAR conference dates.  Most of the city has accessible bike and walking trail connections.  In addition, Edmonton is a 4 hour drive from Banff National Park, Canada’s oldest National Park and Alberta’s most visited tourist destination.  Visitors to Banff in the summer can enjoy hiking, camping, canoeing, cycling, fishing, golfing, kayaking, skateboarding, swimming, walking trails, and relaxing at the hot springs.   The local organizers are Bei Jiang and Linglong Kong.  Details about the meeting will be posted on the WNAR web page as they become available.

2017 WNAR Student Paper Competition

Congratulations to the winners of this year’s student paper competition. The Most Outstanding Written Paper winner was Jiacheng Wu from the University of Washington. The Distinguished Written Paper winners were Qike Li from University of Arizona and Sai Li from Rutgers. The Most Outstanding Oral Presentation winner was Brian Williamson from University of Washington. The Distinguished Oral Presentation winners were Linh Nghiem from Southern Methodist University and Matt Hingham from Oregon State University. The students received their award at the conference banquet.

We give a special thanks to the chair of the student paper competition, Adam Branscom from Oregon State University.  We also thank the team of student paper reviewers and judges for the students’ oral presentations and papers: Kate Crespi from University of California Los Angeles, Ken Newman from United States Fish and Wildlife Services, and Chris Sokra from New Mexico State University.

Figure 1: WNAR student paper competition winners, WNAR President Sarah Emerson, and WNAR student paper competition judges Chris Sokra and Kate Crespi

2018 WNAR Student Paper Competition

WNAR sponsors students who enter the student paper competition. All WNAR-region entrants receive their registration fees and banquet dinner ticket for free. Monetary prizes will be awarded to the best papers in written and oral competitions. Information on the 2018 WNAR Student Paper Competition, registration information, and program details for the meeting will be posted as they become available: We look forward to seeing you there.

Megan Othus


IBS Journal Club – 15 February 2018

The Committee on Communications of the IBS invites you to…

Follow us on Twitter!

Why? To get the latest news about IBS activities as it’s happening!

Join now! It’s easy. Find out how.

Already on Twitter? Our username, or handle, is @IBSstats.

How to join us on Twitter:

  1. Sign up for a Twitter account.
  2. Go to the IBS Twitter page.
  3. Click on the Follow button on the right-hand side of the page.
  4. Sign in, and once again click on the Follow button.
  5. You’re done! Of course, to receive notifications from us on your smart phone, you will need to download the Twitter app.

IBS is on LinkedIn!

Join Our Group, and Get Connected to Colleagues Across the Globe!

Did you know that IBS has a LinkedIn Group for biometrics industry professionals? Network with your colleagues instantaneously! Post discussions to the Group, and get comments/feedback from group members on their perspectives or experiences. A great benefit of this Group is that it’s a very easy and free way to communicate with your colleagues who live all over world. Being connected to the IBS Group will allow you to see other connection possibilities, as well and broaden your professional network.  The possibilities are endless. Join our Group today by visiting and search under Groups for “International Biometric Society”.



19 – 21 February
Introduction to the Joint Modeling of Longitudinal and Survival Data, with Application in R
University of Santiago de Compostela, Spain

25 – 28 March
ENAR Spring Meeting
Atlanta, GA, USA

April 3
A Conference for young researchers
Thesis award: “Prix de thèse Daniel Schwartz”
Paris, France

May 28 – June 1
The French Statistical Society
Paris, France

24 – 27 June
WNAR/IMS meeting
Edmonton, Canada

8 – 13 July
XXIXth International Biometric Conference
Barcelona, Spain

July 28 – August 2
Joint Statistical Modeling
Vancouver, BC, Canada

26 – 30 August
Annual Conference of ISCB and Biennial ASC
Melbourne, Australia

24 – 28 September
Summer School on Advanced Bayesian Methods – part II
Leuven, Belgium

3 – 7 December
Australasian Applied Statistics Conference
Rotorua, New Zealand

17 – 20 December
EMR-IBS conference
Jerusalem, Israel


24 – 27 March
ENAR Spring Meeting
Philadelphia, PA, USA