Download the PDF

President’s Corner

Greetings from Kanloops, British Columbia in Canada, where my partner Jonathan and I are a third of the way through a 15-day cycling trip.  Starting from Vancouver, our first five days have been spent climbing through the beautiful foothills and coastal range. The spectacular scenery will of course come in a few more days when we reach the Rockies.  I’m somewhat nervous about the amount of climbing ahead, but I’ve always wanted to go to Lake Louise and am excited for the adventure!  Along the way there have been some interesting signposts and it was fun to be here on July 1st, which is Canada Day!  But please forgive me if my President’s Corner is a little shorter than usual this time!

In the meantime, there are lots of exciting things happening for IBS.  Front and centre are all the various preparations for the International Biometrics Conference to be held in Seoul, Korea, in July 2020. There is a separate section later in this issue with more detail.  Under the strong and effective leadership of Renato Assunção from the Brazilian Region of IBS, the International Program Committee (IPC) has worked hard to select a strong set of Invited Sessions (from just over 70 submissions).  I am pleased to report that the topics are modern and diverse, not only in terms of breadth of topics covered, but also in terms of regional representation and gender of speakers, organizers and discussants.  I am grateful to the IPC for their hard work!  While they can have a little break for a couple of months, we will need their time, effort and concentration again at the end of the year to evaluate all the contributed paper and poster submissions.   We also have an excellent set of Short Courses planned, thanks to the hard work of Annette Kopp-Schneider and the IBS Education Committee which she chairs.  One new and exciting innovation for IBC2020 is the inclusion of keynote speakers.  A number of us feel that having some high-profile keynote speakers can really enhance the attractiveness of a conference and many modern-day conferences use this format.   I am delighted to report that four big-name speakers have accepted our invitation to deliver keynote addresses at IBC2020. In alphabetical order, they are:

Yoav Benjamini

Dianne Cook

Peter Diggle

David Donoho

 

 

 

 

 

We will be updating you further as all the details regarding lecture titles and timing get pinned down.  But I am just thrilled that we have such a talented and diverse set of keynotes lined up and I can’t wait to hear what each of them have to say!  Yoav will be speaking in a special session organized by the International Statistics Institute (ISI) and where he will receive the 2019/2020 Pearson Prize which “recognizes a contemporary research contribution, published within the last three decades, that has had profound influence on statistical theory, methodology, practice, or applications”. The prize is sponsored by Elsevier B.V and you can find more detail about Yoav’s award here.   As I talked about in my previous President’s Corner, ISI and IBS have a really great partnership whereby we organize a special session and sponsor some young IBS members to attend their World Statistics Congress which is held every other odd year, while they reciprocate for our IBC held every other even year.  I am delighted and honoured on behalf of IBS that ISI has chosen to organize their special session at IBC2020 around the presentation of the Pearson Prize.  Peter Diggle will be speaking about spatial and spatio-temporal statistics which of course has applications in pretty much every area of biometry.   Dianne Cook will be talking about visualisation in a special session being organized in collaboration with KRUG, the Korean R User Group.  I’ve asked David Donoho to speak to us about his provocative paper, 50 Years of Data Science, that has already had so much influence in our community.  I think it will be really exciting for us to launch this new initiative of having keynote speakers at IBC.  I predict it will be a great success.

On a different note but equally important note, work on the new website is also moving along well with the help of Peter Doherty and his team at the International Business Office(IBO),  as well as Jürgen Kübler who chairs the Website Advisory Committee.  The new site will have a fresh new look, though will retain a lot of the same structure as our old site.  A few subpages will need a substantial upgrade.  This is especially true of our History page!   We would love to hear from any IBS members who might have an interest in helping with some of these tasks.   There are, however, two big areas of change and enhancement.   The first enhancement is that the new website will facilitate us developing interactive online communities, which is a specialty offering of the company, Higher Logic,  that will be hosting the new website (see www.higherlogic.com/solutions/associations-online-communities for more detail). While it may be a work in progress for a while, our vision is to create a flexible environment where IBS members can communicate with each other easily, where resources can be shared, and discussions supported.   I am particularly eager to establish a young statistician’s community and to establish an international mentoring program.  Higher Logic offers some interesting facilities, including something called Mentor Match that can be used to link potential mentors and mentees according to their interests.  I would love to hear from our membership with any ideas or suggestions as we embark on this new initiative.  Because there will be the capacity to host multiple different online communities, we will be able to use the resource to support the work of the various IBS committees, as well as Representative Council and the Executive Board.  In my experience, conducting committee business via email can be quite a nightmare that clogs inboxes and makes it hard to find relevant materials.  Having a reliable online space that supports document storage and discussion can really streamline the workload and prevent annoying errors with version control etc.   It is possible that interested Regions may also be able to have a dedicated online community to support communication specifically among their regional members.   Speaking of regions, this brings me to the second area of major enhancement that we are hoping to incorporate into the new website, having to do with the way that new members join, or existing members renew their IBS membership.  As you all know, the IBS is very strongly grounded in our Regions.  Having strong Regions not only enhances our ability to communicate with members but facilitates an international community that can fully respect the nuances of local cultures.  However, when it comes to the process of paying our annual membership dues and updating our contact details, the regional structure does make things complicated.   It is an IBS tradition that except for a small number who join as Members at Large, all of us join IBS via a local region.  Regional Officers collect our membership dues which include the central membership fee, as well as an optional Region-specific fee that is used to support local activities.  Once or twice a year, regions complete a bulk transfer of international membership fees to the business office.   For those of us who live outside the US, paying our dues directly to our local region helps cut extra costs associated with international payments.  It also reinforces our sense of connection to our local region.  But it does create a lot of work both for the regional officers and the IBS business office who must manually reconcile membership lists and process transfers.  As part of the website upgrade, we are working to streamline and perhaps automate some of this.   We are very grateful to the British and Irish Region of IBS who have stepped up and offered to serve as the test case as we do this.  We thought about the possibility of just having everyone pay their membership fees centrally, but the consensus seems to be that that we appreciate and value that strong sense of regional alliance and want to keep it that way.  It is not easy, but we are making good progress in terms of thinking through a process that should work well.  Please stay tuned!

From the Editor

Dear colleagues,

Thank you for the greetings and encouragement received from so many members representing such diverse and varied backgrounds. I am happy to learn that the two propositions, published by us in the first issue of the Biometric Bulletin 2019, were well received. As promised, a new column “Response to Editor” is being added to the Biometric Bulletin and this time includes one response for each of the two propositions. We must appreciate the interest taken by Prof. Abhaya Indrayan from India and Prof. Kuldeep Kumar from Australia in responding to the Editor.

I feel really encouraged to continue inviting responses from the readers – this time on ’Basic theme-II’, as well as continued submissions on ‘Innovations’, as invited in Biometric Bulletin Volume 1 of this year. As already proposed each issue of the Bulletin would have some of my personal thoughts that may be biased perceptions on broad as well specific basic themes. I invite your responses, whether these might be corrections, modifications and additional elaborations or explanations, in brief, not exceeding 300 words. Eligible responses shall be published in the next issue of the Biometric Bulletin. This exercise of mutual but educated dialogue and debate can hopefully go a long way in bringing updated clarity on important topics of mutual scientific interest.

Basic Theme – II 

(Reader’s responses are invited)

It is said that Nature is consistently evolving, and we do have to surrender to the dynamic nature of our Universe in response to many happenings that are beyond our expectation and prediction framework or control. Continuous changes in Nature over time bring uncertainty and variability in each and every sphere of life as well affect the objects and subjects of the various Sciences. Even some of the well-known and tested principles and laws of basic Sciences do fail with test of time. This uncertainty and variability prevalent in nature makes it difficult to satisfy our inner urge of acquiring knowledge of objects and subjects around and surround us and also to decode the mechanism involved during travel from cause to effect. Encountering the uncertainty and variability and thereby understanding their role in rational explanation of the facts becomes the basic feature of all sciences. While we can by no means control or remove the factor of uncertainty, we are capable of measuring it in terms of probability. Simultaneously, hypothetical assumptions of the possibility of any event can’t be ignored, but rather can be recognised through probability, hopefully evidenced based. For example, that the sun will rise tomorrow from the east is certain – the probability of success is 1 or 100%.  This is based on the fact that the earth has rotated consistently and in the same manner and fashion for the millions or billions of years, and every single one of those daily trials has been successful with not even a single failure was registered so far. But it is quite logical and possible that Sun may ditch us by not saying good morning, if not tomorrow, maybe after another million or billion years.

Measurements of uncertainty in terms of probability help in many ways during experimentation and also in making out inferences and decisions with known or minimum interference of the chance or luck factor. However, there are two school of thought. One school always gives credit to and believes in the critical role of chance impacting on each and every event or occurrence, perhaps and even most likely beyond our knowledge. For this school of thought, it is sure that luck factor will always prevail and can never be defeated. In contrast the second school of thinking strongly believes that there is no such thing as a luck or chance factor rather it attributes any shortcoming in our current status to ignorance and/or inability to diagnose or decode the fast and complicated processes leading to any event or occurrence. This School maintains an optimistic viewpoint that as the sciences advance and our level of sensitivity, efficiency and wisdom in making observations and measurement improves, the time will come when we should be able to minimise it or to say goodbye to ‘good luck’. If this school of thought is correct, then we should rest assured in statistical terms a Bayesian approach should gather more strength day-by-day.  However, is it possible that Gaussian thinking will continue to prevail?

Response to the Editor

The Editor has raised a very pertinent question regarding prevailing confusion over the domain of Biometrics and Biostatistics (Biometric Bulletin, 36(1), 2019). The overlap in the use of these terms is indeed substantial and sometimes they are interchangeably used. My views on this issue are as follows. I add Medical Biostatistics to this conundrum for more clarity.

To provide a broader perspective, perhaps it would help to start with a clear understanding of what statistics is about. Although the term is used as plural for various kinds of data, I restrict to its use in singular form. From the epitome of data crunching (Statistics for Big Data for Dummies by Anderson  and Semmelroth, Wiley, 2015 ), the subject of statistics has travelled a long way to now rightly understood as the one that helps in managing uncertainties (A Beginners Guide to Uncertainty of Measurement by Bell, NPL, 1999). To this I must qualify that statistical methods are equipped to handle only the data-based uncertainties, which are commonly seen in all empirical sciences, and other kinds of uncertainties, if any, are excluded. Management is in terms of their measurement of uncertainties by probability and their control by following a proper study design. Statistical methods are a definite help in reaching to a decision in the face of such uncertainties. Although the methods are primarily for uncertainties generated by sampling fluctuations due to intrinsic and man-made variations but other uncertainties in the data, particularly those in epistemic (unknown) domain, remain confounded in most situations. Failure to adequately handle epistemic uncertainties is the primary cause for deprecable comments about Statistics’ credibility as a decision science (Editorial, Basic and Applied Social Psychology, 2015). Errors and biases further complicate the decision process, but I keep those aside for the present communication.

In my opinion, the subject Biometrics is primarily statistical part of biology. It is akin to the other ‘metrics’ sciences such as Econometrics, Psychometrics and Technometrics. Though ‘metrics’ signifies the science of measurement, Biometrics in particular conventionally extends to statistical aspects of all living organisms and does not restrict to measurement. As described by the Editor, it incorporates subjects such as ecology, animal science, and forestry besides the human beings. Some papers recently appearing in the journal Biometrics pertain to these subjects (see, e.g., Scharf et al., Animal measurements, 2019, and Stevenson et al., Aerial survey of animal populations, 2019). Most papers in this journal are on development of new statistical methods for studying biology rather than on application of the existing methods for new results. That, I think, delineates the subject of Biometrics very well. Management of uncertainties remains an integral component in development of these methods.

Biostatistics, on the other hand, has come to establish itself as a statistical science concerned predominantly with the issues of human health and conventionally excludes issues pertaining to the other biological creatures despite the ‘Bio’ suffix. It does include animal experimentation though because such experiments are sometimes considered a precursor to human experimentation. Boston University on its website defines Biostatistics as “application of statistical principles to questions and problems in medicine, public health or biology”, and Washington University School of Public Health website says “Using the tools of statistics, biostatisticians help answering pressing research questions in medicine, biology, and public health”. Both emphasize applications although both include biology within its domain. But I consider Biostatistics as a science exclusive to human beings. It is clear that Biostatistics is primarily application of the existing methods rather than development of new methods. There is hardly any place for theorems and lemmas in Biostatistics – other experts may differ. I consider Medical Statistics a British name for Biostatistics. Whereas American universities mostly have departments of Biostatistics, British universities mostly have departments of Medical Statistics.

Much of the confusion is removed by the term Medical Biostatistics that has come up in the past couple of decades. There are now departments of Medical Biostatistics (e.g., The University of Vermont) and books (e.g., Medical Biostatistics by Indrayan and Sarmukaddam, CRC Press, 2002). This subject can be defined as that part of Biostatistics that is predominantly concerned with promotion, prevention, diagnosis, treatment and prognosis of health conditions in individual human beings and communities. Referring to “Medical + Bio” > “Statistics”, this book explains that Medical Biostatistics is more medical than statistical and seeks to integrate the subject with medicine, away from mathematics. This subject is more focussed to medicine than Medical Statistics has been, and the primary audience is medical and health professionals. The dominant feature of Medical Biostatistics is the management of data-based medical uncertainties in terms of their measurement by probability and control by design as already mentioned for Statistics, but the emphasis now is on medical uncertainties – both aleatoric and epistemic.

In summary, Biometrics is for development of new statistical methods for studying biology, Biostatistics is application of the existing methods to the health issues of individual humans and their population, and Medical Biostatistics is oriented to the issues faced by medical and health professionals. The core of all is the management of data-based uncertainties.

~ Abhaya Indrayan
Biostatistics Consultant, Max Healthcare, New Delhi
Former Professor and Head, Department of Biostatistics and Medical informatics,
Delhi University College of Medical Sciences, Delhi

Response to the Editor’s Proposition 2

Innovations in measurement

Measurement of characteristics like hunger, nutrition, the health of a large society or an individual is always a big challenge. Although lots of work has been done in these areas but still no satisfactory index has been derived, which can be applicable to various scenarios. Statistical techniques like logistic regression and various machine learning techniques like decision tree, random forest, stochastic gradient boosting, Artificial Neural Network, etc. gives the probability of success and failure of the dependent variable (which is binary or categorical in nature) based on certain features (independent variables). This probability between 0 and 1 can be used for ranking individuals, or countries. However, quite often the response variable may not be categorical or there may be no direct response variable is given, but data on various characteristics may be given. In these cases, developing an index or ranking the individuals and countries could be a challenge. One of the techniques which can be used in these cases to develop an index based on factor analysis. In this method, factor analysis is first carried out, and percentage of variance explained by each factor was divided by the total variance explained by the model, then multiplied by each factor score before being summed. This yields a single score for each individual. This single score can then be standardized to give a value between 0 and 100. Details of this approach can be seen in Krishnan (2010) or for more details contact the author at kkumar@bond.edu.au

Krishnan, V. (2010). Constructing an area-based socioeconomic index: A principal components analysis approach. Edmonton, Alberta: Early Child Development Mapping Project.

Professor Kuldeep Kumar, Bond University, Gold Coast, Australia

International Biometric Society 30th International Biometric Conference

Greetings  from IBC2020

On behalf of the 30th International Biometric Conference (2020IBC) Local Organizing Committee and International Biometric Society Korean Region (IBS Korean Region), it is our greatest honor and pleasure to host the 2020 IBC in Seoul, Korea. International Biometric Conferences have been the most effective and prominent gathering of biometric professionals and having the Conference convened in Seoul would be particularly meaningful, to the extent that this highly acclaimed conference would change the face of bioscience in Korea and beyond in every way. Biosciences in Korea and Asia have been recognized as one of the most promising industries for growth (research field for better health and environment), yet there still is a lot of room for improvement.

The 2020 IBC Local Organizing Committee and IBS Korean Region will exert all efforts to design comprehensive and rewarding scientific programs and all participants will have various opportunities to strengthen professional networks and friendship with one another in and around the Conference

Seoul, the heart of the nation and the venue city for the 2020 IBC, is a popular destination for international travelers and where tradition meets modernity in perfect harmony. Seoul also has been home to many exciting and outstanding international conferences and events in the past few decades. The City is safe and tourist-friendly and offers the warmest hospitality, the state-of-the-art conference and comfortable accommodation facilities, breathtaking sceneries, and appealing cuisines.

2020 IBC’s cultural and social programs will be organized for memories to cherish for life and there will also be many other opportunities to explore Seoul and Korea. The 202 IBC Local Organizing Committee and IBS Korean Region, in close association with central and local governments, industries, academic societies and institutions in Korea, assure the success of the 2020 IBC.

IBC 2020 Short Course and Invited Session Offerings are Confirmed

Thank you to all of those who submitted a proposal to present a Short Course or Invited Session during the 30th International Biometric Conference in Seoul Korea.  We are please to announce that the IBC 2020 Short Courses and Invited Sessions have been selected.

Our Education Committee has selected seven short course proposals for presentation just before the International Biometric Conference begins. All Short Courses will take place on Sunday, 5 July 2020 from 9:00 AM – 5:00 PM.

2020 Short Courses

SC01: Targeting learning: bridging machine learning and causality (full day)
Presenters: Antoine Chambaz & David Benkeser

SC02: Model-based Geostatistics for Global Public Health (full day)
Presenters: Peter Diggle & Emanuele Giorgi

SC03: Measuring the Impact of Nonignorable Missing Data (half day)
Presenters: Daniel F. Heitijan, PhD & Hui Xie, PhD

SC04: Data analysis using hierarchical generalized linear models with R (full day)
Presenters: Lars Ronnegard & Youngjo Lee

SC05: Causal questions and principled answers: a guide through the landscape for practicing statisticians (full day)
Presenters: Saskia le Cessie & Els Goetghebeur

SC06: Extension of survival models for correlated data: joint frailty models with recurrent events, a longitudinal biomarker and a terminal event (full day)
Presenters: Virginie Rondeau, Agnieszka Krol

SC07: Topological and Object-Oriented Data Analysis (full day)
Presenters: James Steve Marron, Yuan Wang & Moo K. Chung

In additional to selecting the Short Courses, the Education committee is also in charge of organizing the Statistics in Practice session, which is part of the main scientific program that will run from 6-10 July 2020. The committee picked a proposal that will be especially suited for this purpose:

Statistics in Practice

Fast algorithms and modern visualisations for feature selection
Presenters: Samuel Mueller and Garth Tarr

The International Program Committee (IPC) has selected nineteen (19) Invited Session presentation that will be part of our main scientific program that will run from 6-10 July 2020.  Please note that session speakers and discussants are still being finalized and will be posted in the next issue of the Biometric Bulletin.

IBC2020 Invited Sessions

IS01: Beyond the Proportional Hazards Model: Analysis of Treatment-effects on Time-to-event Endpoints in Cancer Drug Development
Session Chair: Geert Verbeke, Catholic University of Leuven and Hasselt University

IS02: Recent development in methodologies for neuroimaging data
Session Chair: Min Jin Ha, University of Texas MD Anderson Cancer Center

IS03: Recent Advanced Bayesian Methods for Complex Biomedical Data Analysis
Session Chair: Jian Kang, University of Michigan

IS04: Biostatistics and Team Science
Session Chair: Knashawn H. Morales, University of Pennsylvania

IS05: Estimating causal effects in exposures on survival: Methodology in action
Session Chair: Stijn Vansteelandt, Ghent University and London School of Hygiene & Tropical Medicine

IS06: From big to smart data in biostatistics: an integrative approach
Session Chair: Clelia Di Serio, IBS-Italian Region, Vita-Salute San Raffaele University

IS07: New advances in capture-recapture methods
Session Chair: Pere Puig, Universitat Autònoma de Barcelona

IS08: Advanced statistical methods for data arising from High Throughput Phenotyping in Agriculture
Session Chair: Arunas (Ari) Verbyla, Data61, CSIRO

IS09: Innovative Theory and Applications of Hidden Markov Models in the Biosciences
Session Chair: Paul S. Albert, National Cancer Institute

IS10: Statistical issues in developing and evaluating polygenic risk models
Session Chair: Celia Greenwood, Lady Davis Institute for Medical Research, Jewish General Hospital

IS11: Methodological issues in observational studies: promoting best analyses and relevant research
Session Chair: Willi Sauerbrei, Chair, STRATOS initiative, Institute of Medical Biometry and Statistic, University Medical Center Freiburg, Freiburg (Germany)

IS12: Event History Modelling in Register Based Studies
Session Chair: Michael C. Sachs, Department of Medicine, Karolinska Institute

IS13: Data Fusion and Statistical Matching
Session Chair: Saumyadipta Pyne, PhD, Scientific Director, Public Health Dynamics Lab, University of Pittsburgh, Pittsburgh, USA

IS14: Interactive visualization for effective decision-making in agricultural sciences
Session Chair: Professor Dianne Cook, Department of Econometrics and Business Statistics

IS15: Biometrics Research by Women from Around the Globe
Session Chair: Nusrat Jahan, James Madison University

IS16: Exploring the limits of multivariate and high dimensional inference methods to analyze data in ecology
Session Chair: Markus Pauly, Faculty of Statistics, Technical University Dortmund

IS17: Spatial confounding in biometrical applications: ubiquitous undesirable effects and alleviation proposals
Session Chair: Lola Ugarte, Public University of Navarre

IS18: Longitudinal modelling of compositional data
Session Chair: Dr. Hae-Won Uh, Department of Biostatistics and Research Support, Div. Julius Center, UMC Utrecht

IS19: Sparse and dense temporal and spatial data analysis in Health Sciences
Session Chair: Jeanine Houwing-Duistermaat, University of Leeds

Be on the lookout for new exciting updates for IBC 2020!

IBS Awards Update

IBC2018 Travel Awards – A post conference update

The IBS Travel Awards program is made available through the financial support of the International Biometric Society, IBS Regions, and individuals. This award assists IBS members from a Developing Country (DC) to attend an IBC.

Each travel award honoree has been asked to provide a report of their time during the IBC. Over the next few issues of the Biometric Bulletin, you will be able to read their experiences. We hope you enjoy!

Cristian González Prieto, Colombia

My name is Cristian González Prieto, I am from Colombia and I was one of the IBC2018 Travel Award winners, which allowed me to attend the XXIX International Biometric Conference. As a master student, being able to attend this event was really enriching, since it allowed me to update the latest advances in the field of statistics and its large number of applications. In addition, it allowed me to share my work, as well as knowing wonderful people and enjoy the beautiful city of Barcelona. Thanks to the funding provided by the IBS, I was able to fulfill one of my dreams, it also gave me the opportunity to raise more ambitious plans and bigger challenges.

As far as the oral presentation is concerned, exposing my work in another language was one of the biggest tasks that I faced in the presentation of this year. This situation let me to reflect that language should not be a barrier to transmission of knowledge, so that, as a project, I consider to better handle the language for the IBC2020.

Furthermore, the initiative to highlight the work of women who have made such important progress in the area seemed very nice to me. It shows that we are all equal and that knowledge is universal, and anyone can reach it and work on it.

The poster sessions were a space to talk closely with each one of the researchers about their project, and about other projects that are being developed or that they have in mind.

The meeting with the committee was the perfect opportunity to thank for the financial support I was given in order to make the trip and the accommodation in the city. As I mentioned that day, I had never left my country and for me it was a dream come true presenting the preliminary results of something I’d been working on for 2 years: my master’s thesis.

I reiterate my thanks to the organizing committee, to each of the people with whom I had contact and to the IBS for their help. I hope that this type of aid will continue to be given to many people who, like me, would like to attend the event, but, for economic reasons, cannot do it. I also hope to be in Seoul to continue sharing experiences and making friends around the world.

 

Editorial Updates

Biometrics

September 2019 Issue Highlights 

The September issue contains a broad area of Biometric Methodology papers, including “Inference for case-control studies with incident and prevalent cases,” by Marlena Maziarz, Yukun Liu, Jing Qin, and Ruth M. Pfeiffer, “A cluster-adjusted rank-based test for a clinical trial concerning multiple endpoints with application to dietary intervention assessment,” by Wei Zhang, Aiyi Liu, Larry L. Tang, and Qizhai Li, “Fast likelihood-based inference for latent count models using the saddlepoint approximation,” by W. Zhang, M.V. Bravington, and R.M. Fewster, “Causal Inference When Counterfactuals Depend on the Proportion of All Subjects Exposed,” by Caleb H. Miles, Maya Petersen, and Mark J. van der Laan, “An iterative penalized least squares approach to sparse canonical correlation analysis, by Qing Mai and Xin Zhang, “High dimensional mediation analysis with latent variables,” by Andriy Derkach, Ruth M. Pfeiffer, and Ting Huei Chen, and Joshua N. Sampson, “Accounting for phenology in the analysis of animal movement,” by Henry R. Scharf, Mevin B. Hooten, Ryan R. Wilson, George M. Durner, and Todd C. Atwood,  “Granger Mediation Analysis of Multiple Time Series with an Application to fMRI,” by Yi Zhao and Xi Luo, and “Confidence Bands for Multiplicative Hazards Models: Flexible Resampling Approaches,” by Dennis Dobler, Markus Pauly, and Thomas H. Scheike.

The Biometric Practice section features papers spanning a diverse range of application areas, including

“A Sensitivity Analysis Approach for Informative Dropout using Shared Parameter Models,” by Li Su, Qiuju Li, Jessica K. Barrett, and Michael J. Daniels, “Efficient methods for signal detection from correlated adverse events in clinical trials, by Guoqing Diao, Guanghan F. Liu, Donglin Zeng, William Wang, Xianming Tan, Joseph F. Heyse, and Joseph G. Ibrahim, “Measurement error correction and sensitivity analysis in longitudinal dietary intervention studies using an external validation study,” by Juned Siddique, Michael J. Daniels, Raymond J. Carroll, Trivellore E. Raghunathan, Elizabeth A. Stuart, and Laurence S. Freedman, “A Bayesian random partition model for sequential refinement and coagulation,” by Carlos Tadeu Pagani Zanini, Peter Mueller, Yuan Ji, and Fernando A. Quintana, and “Empirical-likelihood-based criteria for model selection on marginal analysis of longitudinal data with dropout missingness,” by Chixiang Chen, Yuan Xue, Lijun Zhang, Biyi Shen, and Ming Wang.

A Biometrics paper won the ASA 2019 Outstanding Statistical Application Award

The paper

Liangyuan Hu, Joseph W. Hogan, Ann W. Mwangi, Abraham Siika (2018). Modeling the Causal Effect of Treatment Initiation Time on Survival: Application to HIV/TB Co-infection. Biometrics, 74, 703–713.

is honored with the American Statistical Association 2019 Outstanding Statistical Application Award. This award, established in 1986, is to recognize the authors of papers that demonstrate an outstanding application of statistics in any substantive field. The award will be announced at the ASA President’s Address and Founders and Fellows Recognition at the 2019 Joint Statistical Meetings in Denver, Colorado, U.S.A. (July 30, 2019).

Note that also in 2018 Biometrics authors received the award. It then went to Hou, Tebbs, Bilder, and McMahan (Biometrics 2017). These authors were the 2017 winners of the Best Paper in Biometrics by an IBS Member recognition, an award from our own Society, bestowed during even-numbered (IBC) years.

2020 – 2022 Co-editor for Biometrics identified

The search committee to identify a successor for Co-Editor Malka Gorfine, whose term will end 31 December 2019, has identified Alan Welsh (Australian National University) as Malka’s successor. Alan will serve from 1 January 2020 to 31 December 2022.  The search committee was composed of the Executive Editor (Geert Molenberghs, Belgian Region), Mark Brewer, Biometrics CE (British and Irish Region), Malka Gorfine, Biometrics CE (EMR), Debashis Ghosh, Biometrics CE (WNAR), Hans-Peter Piepho, Editorial Advisory Committee Chair (German Region), Marcos Malosetti, EAC member (Netherlands Region); Ruth King, EAC member  (British & Irish Region).

Associate Editor Panel turnover

The panel of Associate Editors is composed of colleagues who each serve two-year terms. This means that every 1 July a number of them end their terms. The Co-Editors have invited new Associate Editors to replace colleagues retiring and, in view of the large number of submissions received, also to expand the size of the panel. New Associate Editors are chosen to replace expertise lost by retiring Associate Editors, as well as to fill in gaps in expertise in certain areas, especially in emerging fields. The panel will consist of 96 members, starting 1 July 2019, rather than the 87 before this point in time.

We are grateful for the service, rendered by our retiring Associate Editors: Veerabhadran Baladandayuthapani, Sudipto Banerjee, Pulak Ghosh, Timothy Johnson, Ruth King, Martyn Plummer, Donatello Telesca, and Albert Vexler. We wholeheartedly welcome the colleagues joining the panel: Anabel Forte Deltell, Vanessa Didelez, Ying Guo, Jianhua Hu, Katja Ickstadt, Jian Kang,  Katerina Kechris, Suprateek Kundu, Duncan Lee, Markus Pauly, Sofia Villar, Abdus Wahed, Yuanjia Wang, Xianyang Zhang, Xin Zhang, and Yi-Hui Zhou.

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

The June issue of JABES features nine outstanding papers.  Of particular note is Christopher Wikle’s paper entitled “Comparison of Deep Neural Networks and Deep Hierarchical Models for Spatio-Temporal Data”.  This paper cleverly draws parallels between deep learning and hierarchical models with the aim of demystifying deep learning for those familiar with hierarchical modeling.  The paper also discusses the latest advances in applying deep learning to spatiotemporal data.  Given the emergence of deep learning, this paper is an essential read for statisticians working on environmental and agricultural problems.

The June issue also includes the following papers:  “Efficient Sequential Monte Carlo Algorithms for Integrated Population Models” by Axel Finke, Ruth King, Alexandros Beskos and Petros Dellaportas;  “Multivariate Bioclimatic Indices Modelling: A Coregionalised Approach” by Xavier Barber, David Conesa, Antonio López-Quílez and Javier Morales; “Spatiotemporal Balanced Sampling Design for Longitudinal Area Surveys” by Zhonglei Wang and Zhengyuan Zhu; “Computation of Lacunarity from Covariance of Spatial Binary Maps” by Kassel Hingee, Adrian Baddeley, Peter Caccetta and Gopalan Nair; “Hypothesis Tests for Principal Component Analysis When Variables are Standardized” by Johannes Forkman, Julie Josse and Hans-Peter Piepho; “Post-processing Multiensemble Temperature and Precipitation Forecasts Through an Exchangeable Normal-Gamma Model and Its Tobit Extension” by Marie Courbariaux, Pierre Barbillon , Luc Perreault, and Éric Parent; “Quasi-beta Longitudinal Regression Model Applied to Water Quality Index Data” by Ricardo Rasmussen Petterle, Wagner Hugo Bonat and Cassius Tadeu Scarpin; and “Modelling Spatio-Temporal Variation in Sparse Rainfall Data Using a Hierarchical Bayesian Regression Model” by Sabyasachi Mukhopadhyay , Joseph O. Ogutu, Gundula Bartzke, Holly T. Dublin and Hans-Peter Piepho.

If you have a suggestion for a special issue, we 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 http://link.springer.com/journal/13253. 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 (ken_newman@fws.gov).

Please follow us on Twitter:  @JabesEditor.

Brian Reich
Editor in Chief

Software Corner

A new tidy data structure to support exploration and modeling of temporal data.

By: Earo Wang

Introduction

Going from raw temporal data to model-ready time series objects is painful. Temporal data in the wild can arrive in many possible formats: irregular or multiple time intervals, point events that need aggregating, multiple observational units or repeated measurements on multiple individuals, and heterogeneous data types. For modelling purposes, time series data needs to be coerced into simplified matrices, however there is currently no formal process for how to organize and map wild time series data into model-ready data frames. This results in a myriad of ad hoc (and possibly inaccurate) solutions and the loss of the data richness while in transition. Mining temporal information from time series objects is also inhibited by limited toolkits that hinge on that specialist format. Figure 1 illustrates the lumpy workflow of current time series analysis.

Figure 1 Illustration of the current time series analysis workflow, adapted from R for Data Science (Wickham and Grolemund 2016). The missing “tidy” temporal data structure leads to a myriad of ad hoc solutions and duplicated efforts from raw temporal data to model-oriented time series objects.

 

The tsibble package aims to make this workflow more robust, accurate, and efficient, as visualised in Figure 2. It provides a new data abstraction to represent temporal data, referred to as “tsibble”, allowing the “tidy data” principles (Wickham 2014) to be brought to the time domain. It also helps to lay the plumbing for temporal data analysis modules of transformation, visualisation, and modeling, to achieve rapid iteration in gaining data insights.

Figure 2 Tsibble defines tidy data in temporal context and lubricates the process of time series analysis

 

Tidy data representation for temporal data

First and foremost, tsibble represents tidy temporal data. Contextual semantics—index and key—are introduced to tidy data in order to support more intuitive time-aware manipulations and enlighten new perspectives for time series model inputs. A variable representing time, referred as “index”, provides a contextual basis for temporal data. The “key” consists of one or more variables that uniquely identify the observational units recorded over time. The remaining columns are considered as measurements. Figure 3 demonstrates the shape of a tsibble.

Figure 3 The architecture of the tsibble structure is built on top of the tidy data, with time-series contextual semantics: index and key.

In order to be a valid tsibble we require distinct rows as identified by the index and key. Duplicate rows in the index and key variables signal a data quality issue, which would likely affect subsequent analyses and hence decision making. This requirement ensures the tsibble is a valid input for time series analytics and models.

To create a tsibble, one needs to declare the index and key. Hourly pedestrian counts at different locations in downtown Melbourne, Australia are used as an example. The “key” includes four sensors, and the “index” contains date-times of one-hour interval along with the time zone of Melbourne. The output displays contextually comprehensive details that aid users in understanding their data better.

Domain specific language for wrangling temporal data

Besides a data abstraction, tsibble is a domain specific language (DSL) in R for transforming temporal data. It leverages the tidyverse verbs and adds some new verbs/adverbs to the vocabularies for more intuitive time-based manipulations. Since a tsibble permits time gaps in the index, it is good practice to check and inspect any gaps in time following the creation of a tsibble, in order to prevent inviting some avoidable errors into the analysis. A suite of verbs are provided to understand and tackle implicit missing values; and fill_gaps() turns them into explicit ones along with imputing by values or functions. For example, the code below, sets implicit missing counts in the pedestrian data set to zero before finding the lag and plotting.

A new adverb index_by() adjacent to a verb allows for index subgrouping operations. In conjunction with summarise(), it performs aggregations over time to different time resolutions.  The following snippet demonstrates how monthly average pedestrian counts are computed for each sensor and then plot with ggplot2.

The tsibble DSL provides a set of functional yet evocative verbs for empowering users to compose expressive statements; that facilitates to frame the data analytic problems and document how to approach them through clean code.

Rolling windows with functional programming

Rolling window calculations are widely used techniques in time series analysis, and often apply to other applications. These operations are

Conclusion

The tsibble R package articulates the time series data pipeline, which shepherds raw temporal data through to time series analysis, and plots. It also lays the fundamental computing infrastructure of a new ecosystem for tidy time series analysis, known as “tidyverts”, including time series features (the feasts package) and forecasting (the fable package).

Note

This is a joint work with Dianne Cook and Rob J Hyndman. A longer version of the data structure and design principles about the tsibble R package can be found at arXiv (https://arxiv.org/abs/1901.10257).

References

Wickham, Hadley. 2014. “Tidy Data.” Journal of Statistical Software 59 (10). Foundation for Open Access Statistics: 1–23.

Wickham, Hadley, and Garrett Grolemund. 2016. R for Data Science. O’Reilly Media. http://r4ds.had.co.nz/.

STRengthening Analytical Thinking for Observational Studies (STRATOS): Introducing the Study Design Topic Group (TG5)

Mitchell H. Gail1 and Suzanne Cadarette2

on behalf of STRATOS TG5

1Biostatistics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, USA;  Email: gailm@mail.nih.gov

2Leslie Dan Faculty of Pharmacy and Dalla Lana School of Public Health, University of Toronto, Toronto, Canada; Email: s.cadarette@utoronto.ca

Previous issues of the Bulletin introduced the STRATOS Initiative and activities of Topic Groups TG1 (Missing Data), TG2 (Selection of Variables and Functional Forms in Multivariable Analysis), TG3 (Initial Data Analysis), and TG4 (Measurement Error and Misclassification).  Future issues will cover TG8 (Survival Analysis) and TG9 (High-dimensional Data). Here we describe the purpose and activities of TG5 (Study Design). Current members of TG5 are Suzanne Cadarette and Mitchell Gail (Joint Chairpersons), Gary Collins, Stephen Evans, Neil Pearce, Peggy Sekula, Neus Valveny, Elizabeth Williamson and Mark Woodward. Doug Altman, who died in June 2018 (see Matthews et al., 2018), was an inspirational leader in TG5, and his contributions continue to influence our thinking.

The aim of TG5 is to provide accessible guidance for the design of observational studies. The emphasis is on help for planning observational studies.  This includes defining clear study objectives and then choosing an observational design that meets those objectives, acknowledges practical constraints, and minimizes threats to validity. In addition to providing guidance on the use of classical observational designs, TG5 also aims to investigate new designs for special problems that might arise in collaboration with researchers in other Topic Groups.

TG5 has submitted an introductory level paper on how to select an appropriate observational design for detecting an association of an exposure with disease incidence. This paper defines cohort, nested case-control, case-cohort, population-based case-control and hospital-based case-control studies. The paper then discusses the strengths and weaknesses of the various designs and how to choose an appropriate design from among them, in view of study objectives. Several follow-on papers are being considered to cover how real-world constraints and practicalities affect choice of design; what steps can be taken to minimize threats to validity, such as measurement error and missing data; and what might be done to enhance the generalizability of a planned study. Another topic under consideration is challenges for inference from routinely collected data (such as administrative records) and what design features can ameliorate such challenges.  A paper on designing studies of the associations of prescription medicines with osteoporotic fractures is underway.

In addition to writing expository papers accessible to a wide research readership, TG5 members are collaborating with researchers in other Topic Groups to develop or apply more efficient designs for problems that arise in specific applications. Some projects include: validation studies to gather accurate data that can be used to improve analyses based on electronic health records (with TG4); and steps that promote real time quality control of assay and questionnaire data (with TG3). Other potential projects include: comparing survey calibration with multiple imputation as methods to increase efficiency of 2-phase designs (with TG1), and planning studies of prognostic factors (with TG6). Many of these collaborations were solidified through preliminary discussions during the STRATOS Workshop entitled “Toward a Comprehensive, Integrated Framework for Advanced Statistical Analyses” at the Banff International Research Station in Alberta, Canada, June 2-7, 2019 (http://www.birs.ca/events/2019/5-day-workshops/19w5198).

Several TG5 members have made presentations on behalf of STRATOS TG5. Doug Altman spoke on “Issues in popular designs for observational studies” and Peggy Sekula spoke on “Prognostic studies and the need for guidance” at the Joint Conference on Biometrics & Biopharmaceutical Statistics, August 28-September 1, 2017, Vienna, Austria.  Peggy Sekula spoke on “Design of observational studies and the need for guidance – prognostic studies as an example” at the German Association for Medical Informatics, Biometry and Epidemiology (GMDS), September 17-21, 2017, Oldenburg, Germany. Mitchell Gail spoke on “Potential collaborations between Design Topic Group 5 and other STRATOS Topic Groups” (http://www.birs.ca/events/2019/5-day-workshops/19w5198/videos/watch/201906031032-Gail.html) at the STRATOS Workshop in Banff, Canada, June 2-7, 2019.

References

Matthews,R., Chalmers I., Rothwell P. (2018). Douglas G Altman: statistician, researcher, and driving force behind global initiatives to improve the reliability of health research. BMJ 361:k2588

Region News

Australasian Region (AR)
http://www.biometricsociety.org.au/about.html

Remembering Jeff Wood

It is with great sadness that we report Jeff Wood passed away on 22 April 2019. Jeff was our longest standing member, having joined in 1968, and is sorely missed. In 2013 Jeff received our region’s most prestigious honour, the E.A. (ALF) Cornish award. The citation that accompanied this award is reproduced below.

“Jeff Wood has worked as an applied statistician for nearly 50 years and has been a member of IBS for almost as long. He started at the National Vegetable Research Station in the U.K. in 1966 originally under John Nelder and then under Geoff Freeman, both former presidents of IBS. He moved to Australia in 1973 and worked for the CSIRO for 28 years. Since 2001 he has been working at the Australian National University. His applied work has been in a wide variety of areas, but mostly in ecology, agriculture and horticulture. It has led to many co-authorships and he has published over 120 papers.

He has served the IBS in many capacities, as Council Member from 1986-1993, as member or chair of the Finance Committee from 1987-1998, as Treasurer from 1999-2006, and as a member of the Awards and Education Committee from 1982-1984. More locally he was President of the Australasian Region from 1982-1983.

He has also been involved with other statistical societies. He was on the Committee of the Birmingham Local Group of the Royal Statistical Society from 1969-1973. He was Treasurer of the Canberra branch of the Statistical Society of Australia Inc. from 1987-1988 and President from 1991-1992. He was the national Treasurer from 1989-1992.”

Jeff Wood (right) pictured with Peter Clarke at the 2013 IBS-AR conference in Mandurah.

Emi Tanaka, Social Media Coordinator

IBS-AR member, Emi Tanaka (@statsgen), has taken up the role of Social Media Coordinator. She will be, with the help of regional social media liaisons, better connecting members and increasing the visibility of the society to the general public. A recent activity, during the lead up to International Women’s Day on May 12th, saw IBS female members featured on Twitter and Facebook. Find these and more by searching for hashtag #IBSWomen on twitter or facebook! If your region has yet to nominate a social media liaison or you have any comments or suggestions, please send them to Emi at dr.emi.tanaka@gmail.com.

Biometrics by the Botanic Gardens
Regional Conference, Adelaide, 1-6 December 2019

Make a note in your diary and book the first week in December for coming to Adelaide, South Australia.  Short courses and workshops on a range of topics will be given on Sunday 1st Dec and Monday 2nd Dec, with the main conference running from Tuesday 3rd – Friday 6th Dec. Attuned to the interests of the regional membership, the conference will look at statistics research and applications in medicine and public health, agriculture and environment, genetics, natural sciences and education. Invited speakers include international and Australasian statisticians at various stages of their careers: Marti Anderson, Daniela Bustos-Korts, James Carpenter, Claudia Czado, Joanne De Faveri, Max Moldovan, Blair Robertson and Christopher Wikle.

Early bird registration closes: 19th September 2019.

More information is available on the conference website, www.ausbiometric2019.org.

Save the Dates…

26-28 Nov 2019: New Zealand Statistical Association Conference, Dunedin, New Zealand, www.otago.ac.nz/nzsa

1-6 Dec 2019: IBS-AR Regional Conference, Adelaide, Australia, www.ausbiometric2019.org

30 Nov-4 Dec 2020: Australasian Applied Statistics Conference 2020, Victoria, Australia

Vanessa Cave

Central American Caribbean Region (RCAC)
https://www.biometricsociety.org/region/central-american-caribbean/

The 29th Colombian Statistical Symposium

The 29th Colombian Statistical Symposium will be held in Barranquilla, Atlántico, Colombia, July 15-19, 2019.  Listed below are the names of facilitators who will be leading short courses based on the various  topics (shown in brackets).

Workshop Facilitators (Subject)

Brenda Betancour PhD,  University of Florida, Gainesville – USA. (Bayesian Linkage)

Daniel Peña PhD, Universidad Carlos III, Madrid, España. (Dynamic Factorial Models)

Roger Nelsen PhD, Lewis & Clark College, Portland, USA. (Copula)

Ulrich Rendtel PhD, Freire Universität Berlin, Germany. (Sampling).

Richard Davis PhD, Columbia University, New York, USA. (Time Series).

Victor Leiva PhD, Pontificia Universidad Católica de Vaparaiso, Chile. (Data Science).

At the Symposium there will be 150 oral presentations and 90 poster presentations featuring various aspects of theoretical and applied statistics. Information on the Symposium can be obtained from the website: www.simposioestadistica.unal.edu.co or by sending e-mail to lfgrajalesh@unal.edu.co or  simestadi_fcbog@unal.edu.co.

JAMAICA STATISTICS SYMPOSIUM AND PRE-CONFERENCE WORKSHOP SERIES (JASSYM) 2019

The Fourth Biennial Jamaica Statistics Symposium and Pre-Conference Workshops Series (JASSYM 2019) organised by The Jamaica Statistical Society is set for the October 29 to November 1, 2019 on the University of the West Indies, Mona Campus, in Kingston, Jamaica starting at 8:30 a.m. each day.  The theme for the JASSYM 2019 is “Data-Driven Solutions – The Answer to World Problems”.  

The workshop and symposium activities are aimed at highlighting ways in which the utilisation of statistical methods and tools and/or analysis of data can provide guidance in our search for solutions to world problems on a local and international level.  The programmes will attract participants from local, regional and international organisations.  Thus, we invite you to participate as this will help to create an environment that provides an excellent networking opportunity for the sharing of expertise, experiences and knowledge in the field of statistics.

The first workshop in the Pre-Conference Workshop series, a one-day workshop set for October 29 will feature the Capabilities of SPSS found in MS Excel.  On day two, there will be two half-day workshops, one based on Linear Regression analysis using Stata software and the other an Introduction to Data Management and Analysis using R.  As part of the Symposium programme on October 31 and November 1, there will also be a mini workshop based on Sample Size Calculation for Different Scenarios. This training will build capacity in data management and analysis in Jamaica and the wider Caribbean.

The Symposium will have an interesting line-up of presentations from experts in different fields featuring the application and use of statistical methodology in areas such as data security, production management, education, finance, economics, transportation and health.   To be part of this exciting line-up, please submit your abstract for an oral or poster presentation. The Call for Abstracts can be accessed via the JSS Website or the JSS Facebook Page.  There will be special prizes for best oral and poster presentations in the student and regular participant categories. The deadline for submission of abstracts has been extended to June 21, 2019.

JASSYM 2019 promises to be bigger and  better as we bring together a wide network of experts in the field of data management and analysis  and other specialists who will share expertise and experiences in the utilisation of data to handle problems they encounter in their respective lines of work, that have implications for solutions to local and global challenges.   Join The Jamaica Statistical Society on October 29 to November 1, 2019 in Kingston, Jamaica as we advance efforts to promote the utilisation and development of evidence-based solutions and policies.  For further information please contact the JASSYM 2019 Secretariat by sending e-mail to jssbiennial.rc@gmail.com or by telephone using 1 876 927 2471 extension 227.

Novi Younger

Eastern North American Region (ENAR)
https://www.enar.org/

Thank you to Kathy Hoskins

After 23 years, Kathy Hoskins retired as the ENAR Executive Director in May 2019. ENAR gratefully acknowledges all that Kathy has done over the years.

WebENARs

Be sure to check the ENAR Webinar website for updates regarding the upcoming WebENAR series, as well as for links to past WebENARs and their recordings: http://www.enar.org/education/index.cfm.

Societies Consortium on Sexual Harassment in STEMM

ENAR, along with WNAR and the American Statistical Association, has joined the Societies Consortium on Sexual Harassment in STEMM as one of 54 inaugural member associations. The consortium’s mission is to support academic and professional disciplinary societies in fulfilling their mission-driven roles as standard bearers and standard setters for excellence in science, technology, engineering, mathematics, and medical (STEMM) fields, addressing sexual harassment in all its forms and intersectionalities.

2019 ENAR Spring Meeting, 24-17 March, Philadelphia, PA, USA

The 2019 ENAR Spring Meeting was held 24-27 March in Philadelphia, Pennsylvania. Over 1100 participants enjoyed 650 abstract presentations, including talks at 45 invited sessions, 332 contributed oral presentations, 87 posters, and 39 speed posters. The Junior Biostatisticians in Health Research Workshop was attended by 44 participants from academic biostatistics, statistics, and math departments, as well as research scientists from medical centers and other organizations. Junior investigators learned from senior researchers about publishing, inter-disciplinary collaborations, grant writing, and promotions and leadership, and they participated in a new active learning exercise on giving 2-minute elevator speeches and a roundtable lunch. The Fostering Diversity in Biostatistics Workshop was attended by 75 participants from high school students and undergraduates through graduate students and faculty. Each group participated in different interactive activities, including programming, technical writing, and student recruitment roundtables.

Participants of the Fostering Diversity in Biostatistics Workshop. 2019 ENAR Spring Meeting, Philadelphia, PA, USA.

ENAR extends congratulations to John Van Ryzin Award winner Xin Qiu of Columbia University and all recipients of the Distinguished Student Paper Award (Trambak Banerjee, University of Southern California; Magdalena Bennett, Columbia University; Roland Brown, University of Minnesota; Katrina Devick, Harvard T.H. Chan School of Public Health; Byung-Jun Kim, Virginia Tech; Yujia Li, University of Pittsburgh; Jeremiah Liu, Harvard University; Lan Luo, University of Michigan; Glen McGee, Harvard University; Giovanni Nattino, Ohio State University; Tao Sun, University of Pittsburgh; Zhe Sun, University of Pittsburgh; Lili Wang, University of Michigan; Theodore Westling, University of Washington; Qiwei Wu, University of Missouri; Chong Wu, University of Minnesota; Zhixing Xu, Florida State University; Yishu Xue, University of Connecticut; Yuan Yang, University of Michigan). ENAR also congratulations all Poster Competition winners (Behzad Kianian, Emory University; Viral Panchal, Augusta University; Vivien Yin, Mayo Clinic; Grace Yoon, Texas A&M University; Rong Zhang, University of Pittsburgh).

Distinguished Student Paper Award winners. 2019 ENAR Spring Meeting, Philadelphia, PA, USA.

Distinguished Student Paper Award winners. 2019 ENAR Spring Meeting, Philadelphia, PA, USA.

Poster Award winners. 2019 ENAR Spring Meeting, Philadelphia, PA, USA.

2019 JSM, 27 July to 1 August, Denver, CO, USA

The 2019 Joint Statistical Meetings will be held in Denver, Colorado, USA, from 27 July to 1 August. The theme of the 2019 meeting is “Statistics: Making an Impact.” ENAR has been instrumental in helping to put together an outstanding program and thanks everyone who put forth one of the many proposals for invited and topic-contributed sessions. The 11 invited sessions primarily sponsored by ENAR will cover topics about causal inference in failure time settings, novel clinical trial design and analysis for precision medicine, microbiome data science, early phase trials, statistics education, inference and decision-making in population health, risk prediction evaluation, spatio-temporal statistics, risk stratified prevention, cancer etiology and early detection, and multi-sensor mobile technology for monitoring health behaviors. The 4 topic-contributed and 4 contributed sessions include presentations on large healthcare databases, integrating disparate data sources, massive neuronal datasets, machine learning, and Bayesian methods. The JSM online program can be accessed at http://ww2.amstat.org/meetings/jsm/2019/program.cfm. ENAR extends a huge thank you to Michael Rosenblum of Johns Hopkins University for serving on the Program Committee for the 2019 JSM.

2020 ENAR Spring Meeting, 22-25 March, Nashville, TN, USA

The 2020 ENAR Spring Meeting will take place in Nashville, Tennessee at the newly opened JW Marriott Nashville. The Program Committee is soliciting suggestions for invited paper sessions. Please suggest ideas and potential speakers and/or develop a formal proposal. Proposals on topics that have broad potential scientific impact are particularly encouraged. Invited sessions are 105 minutes long, and different formats are encouraged, such as sessions with four speakers, three speakers plus a discussant, or a panel discussion. The invited session proposals will be selected by the Program Committee, which includes the Program Chair, Associate Chair, 15 ASA representatives of ASA sections, and two ENAR at-large members. One participant may be a speaker/panelist in at most one invited or contributed session. The deadline for submission is 15 June 2019. To informally suggest ideas, topics or names of potential speakers, contact Program Chair Juned Siddique (siddique@northwestern.edu). The Associate Program Chair will be Chenguang Wang and Local Arrangements Chair will be Cindy Chen.

Future Meetings

Stay tuned for information about future meetings, including the 2020 Joint Statistical Meetings to be held in Philadelphia, Pennsylvania, USA from 27 July to 1 August, 2020. ENAR is fortunate to have Jeremy Gaskins of the University of Louisville be our representative to the Program Committee. If you have ideas for the meeting, feel free to contact jeremy.gaskins@louisville.edu. The 2021 ENAR Spring Meeting will be held in Baltimore, Maryland, USA from 14 to 17 March 2021.

Jarcy Zee

Eastern Mediterranean Region (EMR)
http://www2.stat-athens.aueb.gr/~emribs/

New officers for the EMR region

The EMR is pleased to announce the incoming region officers: Yoav Benjamini, President

Philip Reiss, Secretary, and the Country representatives: Denitsa Grigovora (Bulgaria),

Christos Nakas (Greece) Vilda Purutcuoglu (Turkey) and Itai Dattner (Israel)

Annual Meeting of the Israel Statistical Association

On June 6, the Israel Statistical Association held its annual meeting at Bar-Ilan University. Over 300 participants attended the conference, representing academia, government and the private sector.  The meeting opened with words in memory of Prof. Ayala Cohen from the Technion who passed away this year.  Prof. Michael Jordan from University of California, Berkeley delivered the keynote lecture entitled “Statistical Machine Learning: Dynamical, Economic and Stochastic Perspectives “. The organizing committee have put together an interesting and diverse program, with invited sessions covering Official statistics, Theoretical data science, applied data science, Privacy, Statistical analysis of big data, and a special session in collaboration with EMR titled “Prediction and uncertainty quantification”. The organizing committee members were Itai Dattner (Chair), Havi Murad (Poster chair), Anat Sakov (Organization chair), Yosi Rinott and Yury Gubman, Yair Goldberg, Moni Shahar, Jonathan Rosenblatt and Saharon Rosset.

The conference also included a poster session. The best poster was awarded to Tzivel Frostig from Tel-Aviv University with the poster “Post-selection inference following aggregate level testing using summary level data from GWAS”. The second place was awarded to Antonio Castellanos from the Technion with the poster “Silent abandonment: Patience estimation and operational impact”.

During the conference, the Israel Statistical Association annual awards were announced:

The Putter prize was given to Yuli Slavutsky from The Hebrew University of Jerusalem for her MSc thesis “Active adversarial training of classification networks”, under the supervision of Dr. Or Zuk. The Peretz prize was given to Nir Keret from Tel Aviv University for his MSc thesis “Marginalized frailty-based illness-death model: application to the UK-Biobank survival data|”, under the supervision of Prof. Malka Gorfine.

The day after the conference, a special workshop was given by Prof. Michael Jordan at Tel Aviv University titled: “Inference Meets Computation: Dynamical, Stochastic and Economic Perspectives”.

Daniel Nevo

German Region (DR)
http://www.biometrische-gesellschaft.de/

The German region started into the new year with 967 members: 10 honorary members, 867 ordinary members, 44 retirees, 5 supporting members and 41 students. President Werner Branath succeeded Andreas Faldum on 20 March.

Predictive Modelling and Supervised Learning

This continuing education event was taught by Matthias Schmid und Thomas Welchowski, of Institut für Medizinische Biometrie, Informatik und Epidemiologie (IMBIE), Universitätsklinikum Bonn. The course was soon booked out with 25 participants (and a long waiting list) and was held on January 24-25, 2019 in Berlin. Topics were: variable selection methods, components of the error of prognosis and strategies of data driven model development, non-parametric unsupervised learning by recursive partitioning or by random forests, model based learning by boosting, statistical learning with bounded and censored data. Participants’ feedback was extremely positive indicating the need for further courses for these topics (Working Group ‘Continuing Education’: http://www.biometrische-gesellschaft.de/arbeitsgruppen/weiterbildung/seminare-und-workshops.html).

Biometrical Aspects of Genome Analysis XIII 

Universität zu Lübeck’s Maren Vens hosted the annual workshop of working group Population Genetics and Genome Analysis, assissted by Silke Szymczak (Kiel) and Dmitry Drichel (Bonn) on February 21st and 22nd. They invited talks: “Handling of cell-type heterogeneity in epigenome-wide association studies” by Justo Lorenzo Bermejo (Statistical Genetics Group, Institute of Medical Biometry and Informatics, University of Heidelberg) and “Single cell isolation and analysis: Methods and applications” by Stefan Kirsch (Fraunhofer Institute for Toxicology and Experimental Medicine, Regensburg). This joint workshop together with working groups Human Genetics of GMDS and Genetic Epidemiology of DGEpi was attended by 34 participants.

Workshop on Computational Models in Biology and Medicine 2019

In March the joint workshop of the GMDS/IBS-DR working groups “Statistical Methods in Bioinformatics” and “Mathematical Models in Medicine and Biology”, held in Braunschweig, featured keynote talks: Ivo Grosse (Halle): “Phylotranscriptomic Hourglass Patterns of Animal and Plant Development and the Emergence of Biodiversity”, Jan Hasenauer (Bonn): “Mechanistic models of large-scale biochemical reaction networks”, Benjamin Werner (London): “Quantitating somatic evolution in healthy and cancerous human tissues”.

About 50 participants presented talks and posters on diverse topics from bioinformatics, machine learning, multi-omics analysis and disease modelling. Abstracts and slides are available from http://www.biometrische-gesellschaft.de/en/arbeitsgruppen/statistische-methoden-i-d-bioinformatik/workshop2019.html. The workshop was organized by Michael Altenbuchinger (Harvard), Ingmar Glauche (Dresden), Klaus Jung (Hannover), Markus Scholz (Leipzig) with local support by Michael Meyer-Herrmann and Philippe Robert.

DAGStat Conference

The 5th conference of the Deutsche Arbeitsgemeinschaft Statistik (DAGStat, a cconsortium of 14 scientific societies and professional associations) drew about 900 participants into the heart of the Bavarian capital, Munich, from March 18 – 22, 2019. According to the motto “Statistics under one umbrella” the conference was organized as a joint meeting including the 65th annual “Biometrisches Kolloquium” of the IBS-DR and the Spring Meeting of the “Deutsche Statistische Gesellschaft”. Special sessions included the poster reception with beer, the panel discussion “Data Science – Hope or Hype”, the Young Statisticians session followed by the panel discussion “What to do after graduating (Bachelor/Master/PhD)? Your career options in Biostatistics” and a “Junior meets Senior” dinner, 4 shorter (“Statistics Bazaar”), 1 longer (“Statistics in Practice”) and 3 long (“Tutorial”) continuing education sessions, and an evening talk “Data Literacy” geared at the general public. The 34 invited speakers hailed from 13 different countries. As there were more than 400 talks in more than 120 sessions, often nine of them had to be held at the same time. Most sessions (10) had topic „Design of Experiments and Clinical Trials“, 6 sessions each: „Advanced regression modelling“, „Time Series Analysis“, „Survival and Event History Analysis“, 5 session seach: „Statistics in Science, Technology and Industry“, „Computational Statistics and Statistical Software“, „Machine Learning”. Abstracts are visible at https://www.conftool.org/dagstat2019/sessions.php. Local organizers were Göran Kauermann and Helmut Küchenhoff of Ludwig-Maximilians-Universität München.

Participants of DAGStat 2019 take a break in the hall of Ludwig-Maximilans-Universität München.

Young Talents Awards

The Bernd Streitberg Award for the best bachelor thesis went to Jasmin Rühl, Universität Ulm, for “Sample Size Calculation in Time-To-Event Trials with Non-Proportional Hazards Using GESTATE”. The Bernd Streitberg Award for the best master thesis went to Stefanie Krügel, Ludwig-Maximilians-Universität München, for “Statistical Approaches to Characterize and Compare Networks of Microbiome Data”. The Gustav Adolf Lienert Award for young researchers had a tie in the first rank, namely Tobias Mütze (Göttingen) for “Group sequential designs with robust semiparametric recurrent event models” in Statistical Methods in Medical Research, and Philipp Wittenberg (Hamburg) for “A simple signaling rule for variable life-adjusted display derived from an equivalent risk-adjusted CUSUM chart” in Statistics in Medicine.

Award winners with president and vice-president of the German Region. From left to right: Vice-president Werner Brannath, Philipp Wittenberg, Stefanie Krügel, Tobias Mütze, Jasmim Rühl, president Andreas Faldum.

Susanne-Dahms-Medal for Jürgen Kübler

During the 65th Biometrisches Kolloquium of the IBS-DR, Jürgen Kübler (Marburg) was honored by the Susanne-Dahms-Medal for his outstanding contributions to the region. Jürgen Kübler was president and vice-president of the IBS-DR from 2012-2015. Furthermore, he was member of the regions advisory board, helped comment on regulatory guidance, and served the IBS internationally in diverse functions, for example currently in the representative council.

Central European Network (CEN) Meeting 2020

The 66th annual Biometric Colloquium in Berlin September 6 to 11, 2020, will be a meeting of the whole Central European network of IBS that includes the Austro-Swiss Region (ROeS) and the Polish Region (PR). It will be a joint meeting with the German Association for Medical Informatics, Biometry and Epidemiology (GMDS). Conference President Geraldine Rauch of the Charité (Berlin Medical School) heads the Executive Board with the following Representatives: Ulrich Mansmann (GMDS), Werner Brannath (IBS-DR), Martin Posch (IBS-ROeS), Thomasz Burzykowski (CEN). To accomodate GMDS, there will be two Scientific Committee, one for Biometry and one for Medical Information Science.

Upcoming Events

June 27-28, 2019, Summer conference of the working group “Landwirtschaftliches Versuchswesen” in Gatersleben;

July 3-6, 2019, Summer school “Missing Values, and Estimands” in Strobl;

October 17-19, 2019, Summer school “Group sequential and adaptive adaptive Methods” in Lambrecht;

October 28-29, 2019, Academy meets Industry in Ingelheim.

Reinhard Vonthein 

Japanese Region (JR)
http://www.biometrics.gr.jp/

The 2019 Annual Meeting of the Biometric Society of Japan

The 2019 Annual Meeting of the Biometric Society of Japan (BSJ) was held on 16-17 May 2019 at Sysmex Hall in the Kobe University School of Medicine, Kobe, Japan. Two invited sessions on statistical analysis using the cancer registry and a winner of the BSJ Honorary Award were organized. In Japan, the national population-based cancer registry started in 2016. The overview of the cancer registry and researches using cancer registry data were presented.

The other invited session is a new attempt to have a lecture from a prominent professor following Q&A. Prof. Takashi Yanagawa (Kurume University), the winner of the BSJ Honorary Award in 2011, delivered his presentation on his life with biometrics focusing on two research fields: the assessment of the effects of a confounding factor and estimation of incidence and disease rates. There were nine oral and three poster presentations. A tutorial session on functional data analysis was jointly organized with the Japanese Society of Applied Statistics.

The BSJ is pleased to announce that the society conferred the BSJ Honorary Award on Prof. Tosiya Sato (Kyoto University) for his long-term contributions to the society. The society also conferred the Young Biostatistician Award on Dr. Masataka Igeta (Hyogo College of Medicine) and Dr. Tasuku Okui (Kyushu University) for their recent publications on over-dispersed count data and association analysis for gut microbial compositional data, respectively.

The 2019 Japanese Joint Statistical Meeting

The BSJ is one of the six sponsoring organizations of the meeting and the 2019 Japanese Joint Statistical Meeting will be held on 8-12 September at Shiga University in Shiga, Japan. The society is organizing two invited sessions, in one of which the two winners of the Young Biostatistician Award conferred by the society will present their research.

Ikuko Funatogawa

Meetings

8-12 September
The 2019 Japanese Joint Statistical Meeting
Shiga University, Shiga, Japan

Nordic-Baltic Region (NR)
http://ibsnbr.org/index.html

The University of Copenhagen was proud to host the second annual Survival Analysis for Junior Researchers (SafJR) conference outside the UK. This was the 8th annual SAfJR conference. It was a three-day event that was aimed at career-young statisticians with an interest in the application and development of time-to-event analysis and related topics. The conference provided a unique opportunity for participants to present and discuss their work with peers at a similar stage in their careers in a relaxed and friendly environment. The event included a short course on Recurrent Event Analysis by Per Kragh Andersen (University of Copenhagen), talks from keynote speakers Nadine Binder (University of Freiburg), Thomas Alexander Gerds (University of Copenhagen) and Christian Torp-Pedersen (Aalborg University), as well as contributed talks, a poster session and an informal and a formal conference dinner. Detailed information on the program can be found on the website: https://publicifsv.sund.ku.dk/~safjr2019

We, the SafJR 2019 organising team, greatly enjoyed the conference, which took place during beautiful weather in the heart of Copenhagen at the Royal Library. Both keynote and contributed sessions were very inspiring and of impressive quality, especially taking into account that for many speakers this was their first conference. In particular we would like to congratulate the oral presentation award winner Sven Erik Ojavee and the poster presentation award winner Natasha A. Sahr with their excellent contributions. We would also like to thank all of our sponsors, including the Nordic-Baltic Region of the International Biometrics Society, for their kind support.

We are looking forward to next year’s SAfJR conference, which will take place from April 1-3 in Ulm. See their website for more information: https://www.uni-ulm.de/mawi/statistics/upcoming-events/survival-analysis-for-junior-researchers-safjr-2020/

Marita Olsson

Announcements

Network & Inter-Regional Activities Funding Programme Update

In 2016, this funding program was launched to enhance interactions among networks or regions. To date, there have been several activities that were funded through this program as for instance:

  • A joint session of the Spanish, the Eastern Mediterranean and the Italian Region in the framework of the 10th Conference of the Eastern Mediterranean Region in Jerusalem, December 2018.
  • Young Statisticians Showcase (YSS) as part of the XVII Spanish Biometric Conference and the VII Ibero-American Biometric Meeting which brings together the Argentinian, Central American and Caribbean, Chilean and Equatorial Regions in València, June 2019.

This program establishes a funding mechanism for such activities on a more formal and regular basis. The above examples illustrate the types of activities that can be considered, but there are many other possible activities that could be funded under the program, so regions should be creative in their proposals. For instance, inter-regional activities like organizing a symposium or a workshop at a conference being organized by a scientific organization other than the IBS could be funded which would also increase the outreach activities of our society. The guiding principle is that the activity must involve and encourage inter-region interaction. The expectation is that there should be some contribution, maybe in-kind, from the participating regions and some general benefit to the Society at large.

During the 2019 March Board meeting the Executive Board approved new guidelines for the funding program as well as an increase to the programs budget. Details about these changes are below:

Amount: 

The Executive Board allocates $20,000 in total per year to this funding program.  Each activity award is subject to a maximum of $5,000.

Eligibility:   

Applications can be submitted twice a year: by March 31 for activities that are scheduled to start in the second half of the same year (July 1 – December 31) and by September 30 for activities scheduled to start in the first half of the following year (January 1 – June 30).

  • Regions or Networks should send their proposal for a Network or inter-Regional activity to the Taskforce of the Representative Council, using the email below
  • In general, a Region or a Network can apply for only one activity in a single funding round. Deviations from this rule need a particular justification.
  • Regions or Networks must demonstrate how funded activities will reinforce links between regions.

Guidelines for Proposals:

  • In the proposal Regions/Networks should give a detailed explanation of the financial needs, the scientific value, and, of course, the potential benefit to the IBS as a whole.
  • Proposals must be submitted electronically and sent to regions@biometricsociety.org.
  • The Taskforce, with assistance from the IBS International Business Office, will be responsible for managing the review process.
  • The outcome of applications will be sent to the Regions/Networks within a month.
  • Confidentiality of the reviews is assured.
  • Successful proposals must provide a meeting report and photographs, which shall be published in the Biometric Bulletin and shall be used in social media communications. Other benefits for the society (like webinars, videos of key sessions for the conference/workshop website) are welcomed and encouraged.

Conditions & Criteria:

Please note that (a) priority will be given to new activities and (b) regions can apply for repeated activities no more frequently than every other year.

When rank-ordering proposals, consideration will be given to:

  • Merit
    • Activity itself – form new network, network activities, inter-regional activities.
    • Benefit for Society as a whole – webinars, etc.
    • Activities for Young Statisticians
  • Funding
    • Travel expenses and costs for lodging may be reimbursed
    • Conference fees should be an in-kind contribution.

Taskforce of the Representative Council:

The Taskforce will

  • be appointed by the Chair of the Representative Council for a two-year period
  • include 5-6 members of the Representative Council (incl. Chair)
  • be approved by the whole Representative Council

IBS Journal Club

The Education Committee of the International Biometric Society (IBS) is excited to announce it will continue to offer the Journal Club discussions in 2019.

The Journal Club is open to all IBS members free of charge. The primary purpose of the Journal Club, apart from presenting worthy papers in a more public setting, is to widen the scope for understanding these papers and to provide a new networking opportunity for IBS members through a regular internet forum. All sessions are recorded and are available on the IBS website here, http://www.biometricsociety.org/education/video-sessions/. To access the recordings, you must login to your IBS account.

15 August 2019 – Click here to register
Title: Bayesian Analysis of 210pb Dating.
Speaker: Marco Aquino López

10 October 2019 – TBD

12 December 2019 – TBD

2019 Journal Club Recordings:

14 February 2019 – Recording available here.
Title: Flexible variable selection for recovering sparsity in nonadditive nonparametric models
Speaker: Inyoung Kim

11 April 2019 – Recording available here.
Title: Sample Size Determination for GEE Analyses of Stepped Wedge Cluster Randomized Trials
Speaker: Frank Li
Discussant: Linda Harrison

13 June 2019 – Recording available here.
Title: Informative group testing for multiplex assays
Speaker: Christopher Bilder

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 www.linkedin.comand search under Groups for “International Biometric Society”.

Click here to view all IBS, IBS Region and Non-IBS Events and Meetings

If you would like to add your meeting or event to our calendar please send an email to IBS@biometricsociety.org with the following information,

  1. Event Title
  2. Event Description
  3. Event Category
    1. IBS Regional Event
    2. IBS Event
    3. Non-IBS Event
  4. Event Link
  5. Start Date
  6. End Date