November 2019 • Issue #509 AMSTATNEWS The Membership Magazine of the American Statistical Association • http://magazine.amstat.org CSP 2020 Heading to West Coast in February
ALSO: Educational Ambassadors Sought StatFest 2019 There are lots of reasons to JOIN SIAM
More than 14,500 mathematicians, computer scientists, engineers, physicists, and other scientists enjoy the many benefits of belonging Activity Groups to the Society for Industrial and Applied Mathematics. SIAM Algebraic Geometry members are researchers, educators, practitioners, and students Analysis of Partial from more than 100 countries working in industry, laboratories, Differential Equations government, and academia. Applied Mathematics Education Computational Science You are invited to join SIAM and be a part of our and Engineering international and interdisciplinary community. Control and Systems Theory Data Mining and Analytics Discrete Mathematics You’ll experience: Dynamical Systems · Networking opportunities Financial Mathematics · Access to cutting edge research and Engineering · Visibility in the applied mathematics Geometric Design and computational science communities Geosciences · Career resources Imaging Science Life Sciences You’ll get: Linear Algebra · SIAM News and SIAM Review · Discounts on SIAM books, journals, Mathematical Aspects and conferences of Materials Science With name recognition and · Eligibility to join SIAM activity groups Mathematics of Planet Earth worldwide visibility, SIAM is the · SIAM Unwrapped Nonlinear Waves and “ideal platform for promoting applied (member e-newsletter) Coherent Structures mathematics…in academia, industry, · The ability to nominate two students Optimization for free membership government, and sister organizations Orthogonal Polynomials · Eligibility to vote for or become around the world.…SIAM will continue
a SIAM leader and Special Functions to play a leading role in fostering
· Eligibility to nominate or to be Supercomputing collaborations between all users nominated as a SIAM Fellow Uncertainty Quantification of mathematics, from students to “ You’ll help SIAM to: teachers to professional scientists, · Increase awareness of the importance because of the exceptional quality of applied and industrial mathematics of its conferences and publications. · Support outreach to students — René Carmona, Paul M. Wythes ‘55 Professor · Advocate for increased funding of Engineering and Finance, Bendheim Center for research and education for Finance, ORFE, Princeton University JOIN TODAY: www.siam.org/joinsiam join by 1/31/19 and get 25% off your membership when you Enter promo code MBnw19 at checkout. SAVE 25%
SOCIETY for INDUSTRIAL and APPLIED MATHEMATICS 3600 Market Street, 6th Floor, Philadelphia, PA 19104-2688 USA · Email: [email protected] Web: www.siam.org · Phone: +1-215-382-9800 or 1-800-447-7426 (toll free in USA and Canada) · Fax: +1-215-386-79999/17 AMSTATNEWS NOVEMBER 2019 • ISSUE #509
Executive Director Ron Wasserstein: [email protected] Associate Executive Director and Director of Operations features Stephen Porzio: [email protected] 3 President’s Corner Director of Science Policy Steve Pierson: [email protected] 5 November TAS Spans Range of Topics Director of Strategic Initiatives and Outreach 6 Educational Ambassadors Sought Donna LaLonde: [email protected] 7 Diagnosing and Treating Noncommunicable Diseases Director of Education Around the World Rebecca Nichols: [email protected] 10 StatFest 2019: Meant to Be Managing Editor Megan Murphy: [email protected] 11 Changes Coming to NSF’s Social, Behavioral, and Editor and Content Strategist Economic Sciences Directorate Val Nirala: [email protected] 12 A Tour of the International Statistics Scene Production Coordinators/Graphic Designers Olivia Brown: [email protected] Megan Ruyle: [email protected]
Advertising Manager Claudine Donovan: [email protected]
Amstat News welcomes news items and letters from readers on matters of interest to the association and the profession. Address correspondence to Managing Editor, Amstat News, American Statistical Association, 732 North Washington Street, Alexandria VA 22314-1943 USA, or email amstat@ amstat.org. Items must be received by the first day of the preceding month columns to ensure appearance in the next issue (for example, June 1 for the July issue). Material can be sent as a Microsoft Word document, PDF, or within an email. Articles will be edited for space. Accompanying artwork will be accepted in graphics file formats only (.jpg, etc.), minimum 300 dpi. No material in WordPerfect will be accepted. Amstat News (ISSN 0163-9617) is published monthly by the American Statistical Association, 732 North Washington Street, Alexandria VA 22314- 14 STATtr@k 1943 USA. Periodicals postage paid at Alexandria, Virginia, and additional Don’t Just Cook Rice mailing offices. POSTMASTER: Send address changes to Amstat News, 732 North Washington Street, Alexandria VA 22314-1943 USA. Send Canadian STATtr@k is a column in Amstat News and a website geared toward people who address changes to APC, PO Box 503, RPO West Beaver Creek, Rich Hill, are in a statistics program, recently graduated from a statistics program, or recently ON L4B 4R6. Annual subscriptions are $50 per year for nonmembers. Amstat News is the member publication of the ASA. For annual membership rates, entered the job world. To read more articles like this one, visit the website at see www.amstat.org/join or contact ASA Member Services at (888) 231-3473. http://stattrak.amstat.org. If you have suggestions for future articles, or would like to submit an article, please email Megan Murphy, Amstat News managing editor, American Statistical Association at [email protected]. 732 North Washington Street Alexandria, VA 22314–1943 USA (703) 684–1221 ASA GENERAL: [email protected] 16 STATS4GOOD ADDRESS CHANGES: [email protected] Statistics at the Tipping Point: Data for Good in AMSTAT EDITORIAL: [email protected] Environmental Advocacy ADVERTISING: [email protected] WEBSITE: http://magazine.amstat.org This column is written for those interested in learning about the world of Data for Printed in USA © 2019 Good, where statistical analysis is dedicated to good causes that benefit our lives, American Statistical Association our communities, and our world. If you would like to know more or have ideas for articles, contact David Corliss at [email protected].
® The American Statistical Association is the world’s largest community of statisticians. The ASA supports excellence in the development, application, and dissemination of statistical science through meetings, publications, membership services, education, accreditation, and advocacy. Our members serve in industry, government, and academia in more than 90 countries, advancing research and promoting sound statistical practice to inform public policy and improve human welfare. departments 19 meetings Contest to Provide Student Winner with Travel Award CSP 2020 Heading to West Coast in February COMMITTEE MEMBER? 22 education 2019 Educational Ambassadors from Nepal, CHAPTER OR SECTION OFFICER? Mexico Attend JSM What’s Going On in This Graph? Enters Third Year ASA Leader HUB IS HERE FOR YOU member news ASA leaders are ASA members who volunteer in 25 People News some way, primarily as chapter or section officers or 28 Awards and Deadlines committee chairs or members. We appreciate your service and aim to make your volunteer experience 30 Section • Chapter • Committee News easy by providing materials you need in one 34 Calendar of Events convenient location. If you cannot find what you are looking for on the Leader HUB or would like 40 Professional Opportunities additional resources added to this page, please contact us and let us know how we can assist you. Follow us on Twitter www.twitter.com/AmstatNews The HUB is on the ASA Community at https:// community.amstat.org/asaleaderhub/home. Join the ASA Community http://community.amstat.org
Like us on Facebook www.facebook.com/AmstatNews
Follow us on Instagram www.instagram.com/AmstatNews
2 amstat news november 2019 president's corner
Making a Statistical Impact with Text Data
ne of the great pleasures—and benefits— mobile devices are collecting personal fitness data, of belonging to the ASA is the opportunity and users want to know how the data can be associ- to hear about problems on which we can ated with how they feel emotionally and physically. makeO an impact. In this column and the next, I’ll Combinations of machine learning and natural mention some of these, which will include progress language processing algorithms have been applied to on the 2019 initiatives (impact, disinformation, such data, and both offer opportunities for statisti- diversity), pointing us to new challenges to pursue. cians to evaluate their accuracies and uncertainties. One big challenge to which statistics can con- Another example arises in experiments to assess the tribute solutions concerns text data. Surveys and accuracy of eyewitness identification. In such experi- Karen Kafadar questionnaires prefer specific-choice questions (e.g., ments, a “mock eyewitness” (lab participant) views member news “On a scale of 1 to 10, rate …”) to avoid open- a video of a crime, and then sees a lineup of, say, six 25 People News ended responses (e.g., “Other (specify) ….”). But photos of possible perpetrators. Experimenters believe in many areas, text responses may be preferable or strongly that confidence in the identification is related 28 Awards and Deadlines even essential. For example, a physician may ask the to accuracy, so they ask their participants, “How con- patient to describe the level and location of discom- fident are you in your identification?” Participants’ 30 Section Chapter Committee News • • fort or the times when pain is greatest. The data choices are “0% confident,” “20%,” “40%,” “60%,” 34 Calendar of Events from such responses can be important for evaluating “80%,” or “100% confident.” One’s answer of “100%” a treatment remedy under consideration (e.g., under may be someone else’s (like a statistician’s) “60%” (stat- 40 Professional Opportunities what patient conditions is the treatment likely to be isticians are not accustomed to claiming “100% certain “successful,” where “success” also must be quantified on anything). and may depend on the patient). Moreover, such experiments are often conducted In more complicated situations, a clinician must with university psychology students who may have mark on a schematic of the body the location and experience with quantifying their confidence with size of bruises, lacerations, and other physical aber- such a scale, but police officers know the typical rations, which are translated into text (e.g., “upper victim or eyewitness asked to identify a perpetrator right arm”). How do we use these data as features from a lineup will have no clue what “60% confi- for classifying patients into categories of severity of dent” means. So, in real life, the officer can only ask physical condition? the eyewitness to describe in words, “How confi- This problem recently came to my attention dent are you?” The response may depend not only from a professor in our school of nursing who often on factors specific to the crime and eyewitness (e.g., is asked to testify about the features associated with time between the crime and identification process, attempted strangulation of an assault victim. The difference in ages and races between eyewitness forensic nurse’s testimony is important because and perpetrator, the eyewitness’s tendency toward nonlethal strangulation may result in a felony con- conservative versus confident responses), but also viction in most states. If the nurse testifies that the on the officer’s interpretation of the response. One features found on physical examination of the vic- officer might interpret, “I’m pretty sure” as “80% tim are consistent with nonlethal strangulation, the confident,” while another officer might interpret it judge may order the person charged to remain in jail as “90%” or “100% confident.” Further, the quan- until trial. So, these features can be crucial evidence titative translation of that response may depend on to support a conviction (otherwise the perpetrator other comments from the eyewitness during the could be released, free to strike again). Electronic identification process. health records generally have these data issues in How do we extract useful information from multiple formats. such responses so we will know whether to advise The challenges of text data arise with personal police officers to trust or dismiss such responses health and recommender systems also. Increasingly, about “confidence”? We need methods to extract
november 2019 amstat news 3 the relevant information quickly, process text data US adult population use Twitter”). The need for the mathematically, summarize the data, and ultimately spatial origin of the tweets was a further limitation. interpret it in the context of other features. And we The authors note from previous studies that only will need to translate our findings into language 1–2 percent of tweets may contain GPS information others can understand! and that “users who enable geotagging of their tweets Some of the same issues related to processing differ demographically from those who do not.” text data arise in classifying news articles as either Nonetheless, such analyses may provide ideas for genuinely informative, mostly informative apart designing and analyzing future studies. (Incidentally, from unintentional reporting errors, or intentionally representativeness may possibly undermine some inaccurate. Such is the task of the “disinformation” experiments in eyewitness identification. Often, they initiative being co-chaired by former ASA President use online platforms to collect voluntary participants Jessica Utts and Duke Computer Scientist and as their “mock eyewitnesses.” How many of you have Associate Chair Jun Yang. How “robust” are news signed up to participate in an online experiment for stories to misinterpretations? In fact, how do we a modest compensation for your time?) even “measure” degrees of misinterpretations so we On the enormous quantities of data in the life sci- can minimize the errors when the news story is read? ences, Dave Dunson said in an article titled “Boston What “loss functions” or “penalty terms” apply to University to Hold Symposium on Statistics and such misinterpretations of text data? Life Sciences,” published in the October 2019 issue Members of this task force are collecting resourc- of Amstat News, “Statistics has had a fundamental es related to documented instances of “disinforma- impact on this paradigm shift in the way life science tion” and open problems related to the classification is being conducted; there is no use in collecting such of news items from multiple sources. They will be data unless we have reliable and reproducible meth- turning that information into a white paper with a ods for analysis and interpretation. The develop- list of possible research directions ripe for collabora- ment of ‘big data’ statistics has freed up scientists to tion between computer scientists and statisticians. be creative in developing and exploring new sources They also are creating a website of resources pertain- of data.” With those developments come the chal- ing to studies on “fake news,” which will be made lenges of characterizing their accuracy in translation available to ASA members. The translation of find- and minimizing the costs associated with misinter- ings for nonstatisticians is even more critical here, as preting them. it affects not just our research collaborators, but also What other broad areas are ripe for statistical the general public. A subset of the task force is inves- analysis? David Williamson is chairing the Impact tigating mechanisms for effectively educating the Initiative, which includes a “challenge” to our com- public about how to recognize attempts at fake news. munity to propose areas in which statistics is well Social media data are another source of text data, positioned to make a big impact. For example, 30 and the analysis of such data raises additional com- years ago, one might have proposed genomics and plications. In a 2019 American Journal of Public proteomics; 10 years ago, perhaps it might have Health paper, Quynh Nguyen and her colleagues been the application of statistical methods and used geotagged Twitter data to study associations design to validating forensic evidence. Today, per- between sentiments and health behaviors expressed haps “disinformation” or processing text data gener- in the tweets and aggregate county-level health out- ally might make the list. If you offer your proposal comes such as rates of mortality, obesity, and sub- to Impact Initiative committee members, you may stance abuse. see a community of statisticians making an impact Analyses of such data raise issues we need to in this world by working on it in the years ahead. consider. First, Steven Piantadosi and his col- Keep those ideas coming—we look forward to see- leagues warned us in their 1988 American Journal ing them! of Epidemiology paper of “The Ecological Fallacy” about drawing inferences from aggregate data for Editor’s Note: Barry Graubard, David Hoaglin, and individual outcomes. Second, users of social media Jessica Utts contributed to this column. in a given county are not likely to be representative of the county’s population (the authors acknowl- edge “only 23% of all internet users and 20% of the
4 amstat news november 2019 THE AMERICAN STATISTICIAN HIGHLIGHTS November TAS Spans Range of Topics Dan Jeske, The American Statistician Editor
he November 2019 issue of The American Statistician features 11 articles and two let- ters to the editor. TThe General section has three articles. The first looks into tests for trends over time in multinomial probabilities. A test of no trend versus the alterna- tive of a trend for at least one of the categories is discussed. The second article formulates a two-sided test by decomposing the alternative into multiple directional alternatives. The test procedure is a set of rules that maps the rejection decision to a specific alternative. The third article addresses the replica- tion crisis through a Bayesian lens. A Bayesian rep- lication factor developed for a two-group setting is extended to a more general multi-group setting. The Statistical Practice section has two articles. The first is concerned with missing data issues in logistic regression. Nonignorable missing responses are considered and a method to reduce the bias in the estimated regression coefficients is presented. The second paper delves into power and sample size considerations for linear mixed models. Mapping the linear mixed model to a multivariate linear model is modeling effort is to understand whether the aid key to making these calculations more tractable. is being delivered efficiently to the areas that most There are two articles in the Teacher’s Corner. need it. The first provides powerful geometric intuition for An article in the History Corner reviews the his- the tail formula used to compute the expected value tory of industrial statistics and profiles several pio- of a random variable. Several interesting examples neering industrial statistics organizations. are given, and multivariate extensions are discussed. A short technical note in this issue revisits the The second paper is a comparative review of alter- mean value theorem for vector-valued estimating native textbooks that might be used for an under- functions. The article is a follow-up to a previously graduate course in nonparametric statistics. published TAS paper on this topic. An article in the Statistical Computing and The November issue concludes with two letters Graphics section proposes a framework for assessing to the editor. The first addresses a TAS paper that statistical computing tools. The framework could be modeled baseball slugging percentage and the sec- useful for developers of statistical tools and/or edu- ond raises an interesting question about how power cators using statistical tools in their courses. calculations should proceed if a significance thresh- An article in the Interdisciplinary section uses old cutoff (e.g., .05) is to be avoided. spatial generalized linear models to study the alloca- To view the articles, visit www.tandfonline.com/ tion of foreign aid to Malawi. The purpose of the toc/utas20/current or log in at www.amstat.org. n
november 2019 amstat news 5 Left: 2019 educational ambassadors Carlos A. Diaz-Tufinio (Mexico) and Vikash R. Satyal (Nepal) with Drew Baughman, Committee on International Relations in Statistics chair. Middle: Saleha Habibullah, 2018 educational ambassador from Pakistan, participates in a session at JSM. Right: Geert Molenberghs, chair of the ASA Committee on International Relations; Saleha Habibullah, 2018 educational ambassador from Pakistan; Seksan Kiatsupaibul, 2018 educational ambassador from Thailand; and David Williamson, an ASA Board member, meet at the Committee on International Relations in Statistics Meeting during JSM 2018 in Vancouver. Educational Ambassadors Sought
he ASA Committee on International • Provide periodic updates and submit a final Relations in Statistics is seeking qualified report by December 2021 to the ASA through nationals from Ukraine, Democratic the Committee on International Relations in TRepublic of Congo, Uganda, Kazakhstan, Statistics and the ambassador’s home institu- Turkmenistan, and/or Central African Republic to tion. The report should describe how the serve as 2020 Educational Ambassadors. transfer of knowledge was accomplished and In 2020, the ASA will select two educational how the ambassador invited statisticians from ambassadors to receive funding to attend the Joint their country to work with them on the sub- Statistical Meetings and take one or more continu- ject matter of their course. ing education (CE) courses in an emerging area of The ASA will arrange and pay for the selected research. The ambassadors will then return to dis- ambassadors’ airfare, hotel, meeting registration, and seminate the subject matter in their home country CE course registration. The ASA also will reimburse and/or region. the visa application fee and provide a $500 stipend. The application deadline is December 31. Candidates for the ambassadorship must hold Selected ambassadors will be expected to com- a PhD or equivalent degree in statistics or a relat- plete the following by December 2021: ed field and be employed at the level of assistant or associate professor at a recognized university or • Attend the 2020 Joint Statistical Meetings in research organization promoting the teaching and Philadelphia, Pennsylvania, to take CE cours- application of statistics. es in an emerging area of statistics research. Interested candidates should send the following to To view CE courses offered in 2019, search committee chair, Drew Baughman, at [email protected]: for “Continuing Education Course” at ww2. 1. A curriculum vitae amstat.org/meetings/jsm/2019/onlineprogram. 2. A proposal (at most two pages) describing • Return to their home country and, within how the ambassadorship will help disseminate the next year, offer the equivalent of a one- knowledge of statistics in your country and semester master’s-level class with no fewer how the ASA’s expectations set forth above than 10 students on the subject matter of will be implemented upon your return home the CE course and write lecture notes in the language locally deemed most appropriate. 3. A letter from the head of your academic unit Ambassadors are expected to repeat the offer- (e.g., department chair, director, or dean) ing at least once. The committee understands indicating institutional support for your dis- there can be practical or organizational semination plan considerations that make this a challenging requirement, so we are prepared to work with For more information or to become involved, the ambassadors to devise an appropriate plan. contact Baughman. n The ASA will also investigate the possibility of providing further assistance through recorded webinars related to the CE course material. 6 amstat news november 2019 Diagnosing and Treating Noncommunicable Diseases Around the World Kelly H. Zou, Jim Z. Li, and Nikuj (Nik) Sethi
Kelly H. Zou, vice president and head of medical analytics and insights at the Upjohn Division of Pfizer Inc.; Jim Z. Li, senior director of medical analytics and insights at the Upjohn Division of Pfizer Inc.; and Nik Sethi, business technology lead at the Upjohn Division of Pfizer Inc., taught a three-part web- based lecture on generating and harnessing real-world evidence (RWE), health information technol- ogy (HIT), and artificial intelligence (AI). The lecture was jointly sponsored by the Biopharmaceutical, Health Policy Statistics, and Statistical Learning and Data Science sections. Here, Richard Zink from the Biopharmaceutical Section asks the lecturers a few follow-up questions.
Kelly, describe your recent diseases (like heart attacks and Nik: I am the BT lead for work since the inter- stroke), cancer, chronic respira- Upjohn’s research, development, view on ASA Biopharm’s tory diseases (such as chronic and medical functions and a peer Podcast in 2017. obstructed pulmonary disease and of both Jim and Kelly. Upjohn’s My colleagues here and I have asthma), and diabetes. mission is relieving the burden been busy looking for ways to of NCDs with trusted, qual- improve treatment for noncom- Give us a brief overview ity medicines for every patient, municable diseases, known as of your current roles and everywhere. We seek to leverage the NCDs, around the world. how you became inter- our portfolio, experience, and According to the World Health ested in statistics, data expertise to become the trusted Organization (WHO), “noncom- science, and business partner of choice to all stake- municable or chronic diseases are technology. holders committed to improving diseases of long duration and gen- Jim: Kelly and I both work in patient health. We aim to signifi- erally slow progression.” According medical analytics and insights, cantly expand access to our high- to the WHO, there are four major research, development, and medi- quality medicines. Our goal is to NCD categories: cardiovascular cal in Pfizer’s Upjohn venture. treat 225 million new patients
november 2019 amstat news 7
Kelly: Some startling statistics: NCDs are the leading cause of death around the world, kill- ing 41 million people per year, by 2025, and it appears to be an equivalent to 71% of all deaths How can RWD be use- achievable goal to make a global globally, with more than three ful for diagnosing and impact in health care. quarters (32 million) occur- treating NCDs? ring in low- and middle-income Jim: While Kelly is a statistician countries. The devastation of Kelly: For medical research- by training, our career paths took such a “silent killer” increasingly ers serving in industry, govern- us from medicine, health care becomes a global health burden. ment, and academia around the policy, [and] business technol- In the US, cardiovascular diseases world, RWD, big data, and digi- ogy to health care innovation. accounted for 840,678 deaths in tal capabilities and technologies In each step of the way, statis- 2016 (i.e., approximately 1 of must be enabled and elevated tics, data science, and analytics every 3 deaths, bit.ly/2oPYVBl). globally, especially when ran- are not only important tools and In 2010, the global cost of CVD domized controlled trials (RCT) methodological bases, but also is estimated at $863 billion, and become increasingly costly. There infrastructure and enablement it is estimated to rise to $1,044 are tremendous opportunities building. Therefore, our diverse billion in 2030. around the world to gain insights backgrounds and capabilities are through RWE and digital infor- complementary, especially for Nik: Given the prevalence mation outside RCTs. interdisciplinary and collabora- of NCDs globally—includ- tive work in the spirit of being ing emerging markets, Greater Jim: For example, in developing fast, focused, flexible, and fun. China, and developed mar- countries in Asia, Latin America, kets—developing new capabili- Africa, and the Middle East, there Describe some of the ties that are practical and built is a large population that can opportunities and chal- on accessible technologies is key benefit from breakthroughs that lenges of RWE for non- to identifying patients that may change patients’ lives. The expan- communicable diseases, be unknowingly impacted by sion of the middle class, the desire known as the NCDs. an NCD, educating health care for quality medicines and thera- Jim: According to the World providers on treatment options pies, and the rise of NCDs are the Health Organization, there are for NCDs—above and beyond a realities these regions face every four major non-communicable single disease—and working with day. Furthermore, the advances disease categories. Here, the word payers to demonstrate through in information and telecom- noncommunicable means these RWE and technology capabilities munication infrastructures have diseases are not contagious. The how addressing NCD burden also drastically improved, which first is cardiovascular diseases broadly can provide significant has enabled a massive amount of (like heart attacks and stroke). value in the health care system. RWD to be generated from many The second is cancer. The third is These opportunities make our sources outside the RCT frame- chronic respiratory diseases (such work ever more important, par- work. Such data sources include as chronic obstructed pulmonary ticularly in developing countries electronic health records, patient disease and asthma). The fourth in the emerging markets. registries, patient-reported out- is diabetes. comes, and digital apps.
8 amstat news november 2019 Nik: Speaking of opportunities That has a direct impact on me, and formats in databases harder to but at the same time challenges, saving me time and giving me a standardize. These countries also cutting-edge innovation can be more personal relationship. tend to have increased data and pri- EDITOR’S NOTE enabled when data are available From a health care perspec- vacy protections, as well as diverse The views expressed or in aggregate. In this new era of tive, AI is having a profound data formats sitting in patient reg- here are the data as precious as oil, data inte- impact across the spectrum, istries without being standardized authors’ and do not gration, access, interoperability, from research and drug develop- and connected (see www.bmj.com/ necessarily represent standardization, quality control, ment to patient care—including content/360/bmj.j5262). those of Pfizer Inc. security, and privacy protection health care management. A great MORE ONLINE are important issues where data example is in medical imaging Kelly: Let me give some final Kelly Zou was is available. A major barrier, how- and diagnostics, in which tech- closing remarks here. It is cer- interviewed for ASA ever, is simply accessing data due nologies are being used to detect tainly an exciting era for stat- Biopharm’s Podcast to a diverse set of sources and skin cancer from photos taken isticians and data scientists to in 2017. Listen to it repositories of data. As we move by patients (see www.skinvision. make meaningful and impactful at www.buzzsprout. com/16296/567031- into a new era of RWD in the com). The key question for lead- contributions to combat these episode-45-kelly-zou. health care landscape, a key chal- ers in the industry is how to put chronic and debilitating NCDs. lenge is going to be how to make such a complex set of informa- Skillful statistics and data sci- data accessible to the right par- tion together within the unique ence capabilities are critical in ties at the right time to only be country-specific health care gaining insights from RWE. For used as prescribed. Finally, once systems and data repositories, digital and technological innova- data is accessible, deriving algo- which will improve diagnoses tions that can target therapies and rithms that are robust and tai- and treatment. optimal treatment strategies, it is lored toward the specific patient imperative to foster collabora- populations will enable patient Jim: The different languages in tions and partnerships. n identification and help address those countries may make variables challenges around the patient journey. The challenge we are facing as an industry is the short- age of talent in both data transla- tors, who are like shepherds with subject-matter expert knowledge, and data scientists. ASA, in cooperation with the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA) under a grant from the National Science Foundation (NSF) is pleased to announce a Senior AI is a commonly used Research Fellow Program for 2020. term these days. Most statisticians likely think The Fellowship Program at BLS allows research fellows to come to the BLS and use BLS data and of it in terms of predictive facilities and interact with BLS staff . More information is available on the BLS website at http://www. bls.gov/osmr/asa_nsf_bls_fellowship_info.htm or in our brochure at https://www.amstat.org/asa/fi les/ modeling. What other pdfs/YCR-ASANSFBLSFellowshipProgram.pdf uses of AI are there? Are Application Deadline: January 3, 2020 there examples of AI that are being used in The Fellowship Program at BEA off ers a unique opportunity to perform research at the Bureau of practice? What does the Economic Analysis. BEA produces key economic statistics that infl uence government policy, fore- future hold for AI? casting and business investment. Fellows will have access to BEA data and the expertise of BEA staff . More information is available at: Nik: From a consumer’s perspec- https://www.bea.gov/research/fellowship-program or in our brochure at tive, AI has in many ways become http://www.amstat.org/asa/fi les/pdfs/YCR-BEA.pdf pervasive in enabling mobile Application Deadline: January 3, 2020 devices and applications running on the mobile devices to be more Eligibility An academically recognized research record and considerable expertise in the area of proposed re- intelligent. For instance, I recent- search required. U.S. government employees are not eligible to apply. Applicants must be affi liated ly got a text message from a num- with a U.S. institution. ber that was not saved in my con- Condition of Appointment/Benefi ts tact list, and the phone (searching Research will be conducted at the government agency. The stipend received is commensurate with through my other messages and qualifi cations and experience. Term of appointment is fl exible. Fringe benefi ts and travel allowances emails) suggested who the person are negotiable. could have been and was correct.
november 2019 amstat news 9 Students read materials from exhibitors during StatFest 2019. StatFest 2019: Meant to Be StatFest 2019 Planning Committee
ropical Storm Imelda Science Center at Houston dropped more than 30 (UTH). Participants from the inches of rain in south- Austin campus joined remotely Teast Texas September 17–19. via the institution’s interactive Conference organizers strongly television connection. considered canceling StatFest Despite Imelda and poor 2019 due to widespread flooding weather conditions, more than and Federal Aviation 120 participants attended: Students listen to Eric Boerwinkle welcome them to Administration–mandated flight approximately 44 percent under- StatFest 2019. stoppages. But on September 20, graduate students; 19 percent the grey clouds began to roll graduate students; and 37 percent away and the high water began to professionals in academia, govern- recede. Planning committee ment, or industry. chair, Adrian Coles, said, “The The program started with a water receded much faster than welcome and opening remarks expected! It’s almost as if StatFest from Eric Boerwinkle, dean 2019 was meant to be.” of the UTH School of Public On September 21, the ASA Health, and Rob Santos, 2021 Committee on Minorities in president of the American Statistics (CMS) held the 19th Statistical Association. annual StatFest conference. Renee Moore of Emory StatFest is a one-day event aimed University Rollins School of at encouraging undergraduate Public Health was this year’s students from historically under- keynote speaker, and Hadley represented groups to consider Wickham of RStudio was a fea- graduate studies and careers in tured speaker. statistics and data science. Two panel discussions intro- Exhibitors connect students to opportunities during This year’s event was hosted duced participants to diverse sta- StatFest 2019. by the University of Texas Health tistics and data science careers in
10 amstat news november 2019 academia, government, industry, and consulting. Folefac Atem of UTH Dallas shared best practices on preparing for graduate school, and Jemar Bather of Harvard T.H. Chan School of Public Health provided practical guid- ance on networking and intrap- ersonal skills. There was a student-only session, titled “The Graduate Changes Coming to Student Experience,” that con- sisted of panelists who are enrolled in master’s and doctoral NSF’s Social, Behavioral, programs in statistics and data science. During the session, fac- ulty and professionals attended a and Economic Sciences parallel session that explored ways to help the represented organiza- tions and institutions become Directorate more diverse and inclusive. Other highlights included he Social, Behavioral, According to Lupia, NSF will MORE ONLINE a poster session, an ice cream and Economic Sciences release a letter explaining the The COSSA webchat social, and an exhibition area. Directorate (SBE) at the changes and include descriptions is available at bit. Throughout the day, partici- TNational Science Foundation of the new programs. In addition, ly/2MpyXh2. pants were able to connect with (NSF) will be releasing changes the directorate will hold a series Slides of the webchat representatives from more than to several of its longstanding of webinars to answer questions can be viewed at bit. 10 graduate programs and pro- programs in the coming weeks. from the social science communi- ly/2MqRb1Q. fessional organizations. During the Consortium ty about the repositioning effort. Although attendance was of Social Science Associations COSSA will share the details of approximately 20 percent less (COSSA) Headlines web- both the letter and webinar series than expected due to Imelda, chat (see bit.ly/2MpyXh2), when they are made available. organizers received positive feed- Arthur Lupia—assistant direc- Of the eight existing pro- back from attendees during and tor for Social, Behavioral, and grams to be changed, the direc- after the event—suggesting #19 Economic Sciences—gave a torate has only released full was another impactful event in preview of what those changes details of the changes to the the history of StatFest. will entail. political science program, which StatFest 2020 is scheduled to The changes, which will go will be split into “Accountable be held at Eli Lilly and Company into effect for deadlines and Institutions and Behavior” (bit. in Indianapolis, Indiana, and solicitation due dates after ly/2Bm1RZ8) and “Security and StatFest 2021 is scheduled to be January 1, 2020, are intended Preparedness” (bit.ly/2oEjUqR). held at Columbia University in to help policymakers and mem- For more information or New York. bers of the public better under- to sign up for additional webi- Read more about StatFest stand the value of the research nars, visit www.cossa.org. Follow 2019 on the ASA’s Committee on funded by SBE. However, the #COSSAHeadlines on Twitter Minorities in Statistics webpage repositioning is not intended to for details. n at https://community.amstat.org/ dramatically alter the types of cmis/events/statfest/statfest-2019. n research SBE supports.
november 2019 amstat news 11 A Tour of the International Statistics Scene Jana Asher, Committee on International Relations in Statistics Member and Slippery Rock University Assistant Professor
s the world’s largest community of statisticians, the home and, within the next year, teaches the subject matter American Statistical Association is an amazingly of the CE course(s) taken to a class of no fewer than 10 diverse body of professionals. Between the sec- students at the master’s-degree level. The EA then sub- tions,A which focus on particular statistical subfields; the mits a written report to the Committee on International chapters, which provide a regional platform for commu- Relations in Statistics about their experience in the pro- nication and collaboration; and the committees, which gram. Finally, during the following academic year, the EA focus on specific ASA initiatives or outreach efforts—not repeats the class he or she taught at least once, either in to mention the in-person meetings and online communi- person or through webinars. ties—it is easy to believe pretty much everything going on The EA program allows ASA members to learn about in the statistics profession is represented here. the statistics scene in a few specific countries each year. The world of statistics, however, extends far beyond the However, there is also an internationally focused commu- membership of the ASA—a fact that has been recognized nity of statisticians from across the world that connects fre- by the ASA leadership for a long time. To make sure the quently through international professional societies. ASA membership is aware of what happens in the statistics profession in other parts of the world, the ASA formed the International Statistics Professional Societies Committee on International Relations in Statistics (CIR) in There are multiple international organizations of inter- 1979. Briefly, the mission of the CIR is: est to statisticians and data scientists, the oldest of which is the International Statistical Institute (ISI, www.imstat. To identify goals, develop policies, and plan projects that org). Elected membership in the ISI is prestigious and foster activities involving international communications, requires letters of recommendation from three cur- exchange, and joint development with professional statisti- rent members; regular membership is also available. cians, societies, and other organizations in countries through- The ISI produces several publications—including the out the world; to keep ASA members and others informed of International Statistical Review—and serves as an umbrella these activities and to foster their involvement in them. organization for the Bernoulli Society for Mathematical Statistics and Probability (bernoulli-society.org), Connecting US Statisticians to the World International Association for Official Statistics (iaos-isi. The CIR helps the ASA membership interact with the org), International Association for Statistical Computing international community mainly through the Educational (iasc-isi.org), International Association for Statistical Ambassador Program (EAP). The CIR launched the EAP Education (iase-web.org), International Association of in 2005 with the following goals: Survey Statisticians (isi-iass.org), International Society • Welcome new international members to the for Business and Industrial Statistics (isbis-isi.org), and American Statistical Association International Environmetrics Society (environmetrics.org). Much like sections of the American Statistical Association, • Create trust between the ASA and other national each of these organizations holds meetings at a big ISI and international statistical societies convention called the World Statistics Congress that • Advance lasting collaboration between the ASA is held once every two years in a different country. The and those societies to allow ongoing exchanges next World Statistics Congress will be in the Hague (The of knowledge Netherlands), July 11–15, 2021. You might have noticed there isn’t a Bayesian Association Every year, the CIR recruits a number of Educational under the ISI umbrella. The International Society for Ambassadors (EAs) from foreign countries and subsidizes Bayesian Analysis (ISBA, bayesian.org) is a separate orga- their attendance at the Joint Statistical Meetings. It is the nization, founded in 1992, that focuses specifically on CIR’s hope that ASA members will interact with the EAs Bayesian analysis techniques. There are chapters of the at the Joint Statistical Meetings, learning more about the ISBA in Australia/Oceania, Brazil, Chile, East Asia, India, statistical community in their home countries. and South Africa; ISBA also boasts several topical sections, At JSM, the chosen EAs attend a continuing education including the Section on Biostatistics and Pharmaceutical (CE) course or courses in an emerging area of research Statistics and Section on Economics, Finance, and Business. (unknown in their own country) and receive ASA full The ISBA holds world meetings every two years; in 2022, membership for one year. Each EA then returns to their they will be in Montréal (Canada) and, in 2024, they will
12 amstat news november 2019 be in Venice (Italy). ISBA also produces the journal Bayesian Finally, the Caucus for Women in Statistics (cwstat.org) Analysis (projecteuclid.org/euclid.ba). deserves to be mentioned here, as the organization has recent- If biostatistics is your focus, you might want to check ly developed a more international focus. Founded in 1971 to out the International Biometric Society (IBS, biometric- address discrimination against women in the statistics field, society.org). Unlike the ISI, the IBS operates primarily the organization was instrumental in the establishment of the through its regional sections, some of which are country ASA Committee on Women in Statistics in the same year. based (e.g., Italian, Polish, Japanese) and some of which The CWS provides multiple benefits to members, includ- are larger (e.g., North African). In the United States, there ing access to webinars, mentoring opportunities, access to are two regional sections: the Eastern North American travel awards, career resources, and a network of committed Region (ENAR, enar.org), which includes all the United women and men who will support your career goals. States and Canada roughly east of Denver, Colorado, and the Western North American Region (WNAR, wnar.org), Regional International Organizations which includes the rest of the US and Canada. ENAR Several regions of the world have inter-country organizations holds annual spring meetings, and WNAR holds an annual that allow groups of countries to pool resources. There are conference in June. IBS, like ISI and ISBA, holds a large multiple professional organizations for statisticians in Europe world conference every two years; the next one will be in that are Europe-wide. For example, the International Society July of 2020 in Seoul (Korea). IBS also produces several for Clinical Biostatistics (www.iscb.info) runs conferences top-quality peer-reviewed publications, including Biometrics annually in different countries within the European Union. (bit.ly/2nTGJpX) and The Journal of Agricultural, Biological, Other Europe-based organizations include Statisticians in and Environmental Statistics (bit.ly/2VQdhxw). the Pharmaceutical Industry (psiweb.org) and the Statistical Maybe you would prefer to become involved with a Modelling Society (statmod.org). more field-specific organization. There are two that also fall There is a regional organization for Africa called the under biostatistics: The Society for Clinical Trials (sctweb. African Statistical Association (afsa-assoc.org); this organiza- org) and the International Genetic Epidemiology Society tion had gone defunct but was recently revitalized through (www.geneticepi.org). Both societies focus on specific appli- the efforts of the Ethiopian Statistical Organization. A recent cations of biostatistics, both have an annual meeting, and development in this geographic area was the founding of the both have an associated peer-reviewed journal. The Society West African Young Statisticians Association (waysas.org); this for Clinical Trials will be meeting in Baltimore, Maryland, organization is also based in Nigeria but provides mentoring in May 2020, and the International Genetic Epidemiology and training to statisticians throughout western Africa. Society is meeting in Hong Kong in July of 2020 (just prior to the IBS meeting in Seoul, South Korea, that year). Summary All these international societies serve similar functions This is a small taste of what is out there to explore in the for their members (i.e., organize meetings, publish journals, worldwide statistical community. provide training opportunities, and generally build a profes- If you are already planning to travel to a different sional community at an international level). The Casualty region of the world, the ASA Committee on International Actuarial Society (CAS, www.casact.org) takes on an addi- Relations in Statistics encourages you to consider attending tional role; they are a credentialing organization for actuar- a local conference or contacting the local statistical organiza- ies. As is stated on their website, CAS members are experts tion to learn where your foreign colleagues are working and in property and casualty insurance, reinsurance, finance, what they are working on. The Committee on International risk management, and enterprise risk management. If you Relations in Statistics will be unveiling a list of national are curious about what is required of an actuary, take a look. statistics organizations from around the world shortly. In In the Asian region, multiple countries have interna- the meantime, if you are traveling and want to be put into tional statistical associations, including China (International contact with a local statistician, contact Jana Asher at Jana. Chinese Statistical Association, ICSA, www.icsa.org/icsa), [email protected]. India (International Indian Statistical Association, IISA, Even if you are not planning to travel anytime soon to a intindstat.org), and South Korea (Korean International different region of the world, the CIR invites you to assist us Statistical Society, statkiss.org). Luckily for monolingual in welcoming future Educational Ambassadors who attend Americans, each of these websites is in English or has an the Joint Statistical Meetings. To start the connection, English version. The International Chinese Statistical Educational Ambassadors will be encouraged to directly Association and International Indian Statistical Association inform the ASA community about the statistical work co-sponsor the Joint Statistical Meetings and hold their own being done within their own country through an article in annual meetings. ICSA’s flagship journal is Statistica Sinica Amstat News prior to their attendance at the Joint Statistical (www3.stat.sinica.edu.tw/statistica), and while IISA’s flagship Meetings. So, look for information about the EAP in future journal for many years was Statistical Methodology (journals. editions of Amstat News and come meet us and our EAs at elsevier.com/statistical-methodology), the organization is in the the Joint Statistical Meetings in Philadelphia next year. n process of developing a new journal and publishes an edited book series (springer.com/series/16210).
november 2019 amstat news 13 columns
STATtr@k Don’t Just Cook Rice
n Chinese, the literal transla- way, don’t be the colleague who Health care is obviously tion for cooking, “zhŭ fàn,” is responds to a request with a table complicated, so there are many “cook rice.” What does this of numbers or data chart. First aspects involved with answer- haveI to do with data science? It’s ask what problem your colleague ing this question. One needs to a useful metaphor for working as is trying to solve. understand how a hospital func- an applied statistician (i.e., data What does it mean to solve tions, how clients make decisions scientist). To explain, let’s take a problem? I’ll use my current (they want good, not perfect, the cooking analogy further and experience to explain. My com- solutions), how data is structured Iyue Sung is compare ourselves to chefs. pany collects—for our clients— in the databases, statistics, pro- vice president of A chef uses both technical data that reflects how well health gramming, and process. The last enterprise analytics skills (e.g., knife skills, cooking care is delivered in hospitals, two are important because we try at Press Ganey, a technique) and subject knowledge medical offices, etc. We measure to scale all the work we do (i.e., health care data, consulting, and (the chemistry and nuance of how aspects such as patient experience develop a process in which we can technology company. ingredients interplay) to develop (perception of different facets of answer similar questions—for all He manages a team and cook a dish. Furthermore, their stay), employee engagement clients—without re-inventing the of data scientists who they don’t think about just the (perception of a hospital’s cul- wheel). The chef wants a kitchen uncover patterns individual dish, but the whole ture, processes, and quality), and running like an efficient assem- related to patient meal. You go to a restaurant not experience, workforce clinical outcomes (e.g., infection bly line to turn out quality meals engagement, nursing to eat, but to experience. rates, readmission rates). Clients consistently while accommodat- quality, and clinical Similarly, a data scientist uses access this data and related ana- ing special requests. quality. Iyue’s prior both technical skills (program- lytics through a web application Putting together the parts experience includes ming to create data visualizations built by our development team. above requires collaboration with electronic health and build models) and subject The obvious purpose of this a range of people—clients, client records company athenahealth and expertise (the math behind statis- information is to improve per- liaisons, health care experts, and Oliver Wyman, a tical methods) to turn data into formance (e.g., reduce infection database architects (to understand strategy consulting knowledge. But your job isn’t rates or decrease nurse turnover). the database’s structure). You may firm. He holds a to just build a model or create a And one of the data science also need to consult your data sci- PhD in statistics cool data visualization. It’s to help team’s function is to help our cli- ence colleague who wrote an R from The Ohio State someone make data-driven deci- ents improve their performance, (or Python) package that can be University and a BA in mathematics and sions. The analysis is one com- beyond what they can do with used for your problem. In other philosophy from ponent of the whole. It provides the application. A common prob- words, a project like this requires Boston University. supporting evidence for the larger lem might be, “I need to improve taking a broad perspective of what You can reach him objective of understanding the the patient-clinician relationship. it means to be a data scientist. by email (iyue.sung@ client’s problem and providing What components of care deliv- To understand this perspec- pressganey.com) or Twitter (@iyuesung). practical solutions. Put another ery impede that relationship?” tive and how our team solves
14 amstat news november 2019 columns
staff enable all patrons to enjoy the experience consistently. Your data science team should have the same components: infra- structure (e.g., database and soft- ware like R); processes (e.g., your own R packages and a version control system); and a wide range of expertise (programming, data engineering, statistics, industry knowledge). You cannot be an expert at everything, but the team can. If your team is well run, you can focus on your project, which is the “meal” of analysis, presen- tation, and decision-making. Through this collaborative expe- rience, you and your colleagues will learn from each other. Art by Iyue Sung What I’ve described isn’t original and can be applied to these problems, I find it helpful much of the day producing ana- most fields. But given you’re an to group our responsibilities into lytic work and less time interact- applied statistician—someone the following four categories: ing with colleagues outside the who savors discovery and profes- sional growth—you’ll surely find • Technical: Programming and team (but you should certainly statistical analysis tasks ask). In contrast, this ratio flips joy in thinking more broadly for group leaders. They spend about what it means to “do data • Process: Organized systems less time on technical matters and science.” Once you’ve mastered to get things done more on nontechnical tasks to cooking just rice (risotto ain’t easy), there is much to learn • Industry: Understanding the find more “business” for the team, about the many ways it can com- big picture of what you’re which, consequently, provides doing them more opportunities to grow. plete a meal and, ultimately, the Going back to the food anal- entire process of providing a con- • People: Interpersonal skills to ogy, think of the data science sistent dining experience. Enjoy get things done team as the restaurant staff. A the experience! n Within each category, a data restaurant requires infrastruc- scientist’s expertise and responsi- ture, processes, and a range bility grow—or shrink—as their of positions to run effectively. career progresses (represented in Together, the chef, manager, the diagram). Initially, they spend cooks, suppliers, and service
november 2019 amstat news 15 columns
STATS4GOOD Statistics at the Tipping Point: Data for Good in Environmental Advocacy nvironmental dangers are Hackathon Focuses on of greenhouse gas emissions in often described in terms of Climate Change Boston. Teams participating in tipping points—a maxi- The C40 Climate Leadership the hackathon developed innova- Emum greenhouse gas level, the Group (www.c40.org) is a net- tive solutions to help understand, total human population our work of major cities around the measure, and limit greenhouse gas planet can support, or the world working together to fight cli- sources. Many of these creative increase in temperature triggering mate change. C40 brings together data science applications can be With a PhD applied to other cities. in statistical a large rise in sea levels. As much mayors, city departments, business astrophysics, David as we hear about tipping points executives, and thought leaders The C40 hackathon is a great Corliss leads a data for environmental disasters, the around a mission to “collaborate example of an event using sta- science team at Fiat concept itself is benign: a mathe- effectively; share knowledge; and tistical science to fight climate Chrysler. He is the matical threshold in a small value drive meaningful, measurable, change, showing many best founder of Peace- Work, a volunteer that, when crossed, leads to a and sustainable action on climate practices needed to host a high- cooperative of large change in something else. change.” Ninety-four cities with quality event that makes the most statisticians and data Tipping points can also be a combined population of more impact: a core group or local scientists providing for good things, like the number than 700 million people partici- volunteers collaborated with a analytic support for of trees saved and planted for a pate in the network. A particular larger organization championing charitable groups sustainable oxygen supply or the focus of C40 is practical, substan- the cause they want to promote. and applying statistical methods number of statisticians partici- tive, and effective implementa- A small core group of organiz- in issue-driven pating in a hackathon to form a tion of the 2016 Paris Agreement ers developed careful plans and advocacy. nexus of cross-pollinating creative on climate change. carried the project through. A ideas for a better world. Last year, C40 partnered with trusted organization provided Examples of statistics at the Qlik and the city of Boston, a C40 data that can be easily transferred tipping point are all around us, member, to sponsor a hackathon. to the laptops of hackathon par- and all of us can participate in Boston supplied data energy use ticipants. A centrally located site making a difference for good. in buildings across the city, which with parking available nearby Here are some examples. accounts for almost three-fourths facilitated participation, and
16 amstat news november 2019
NOVEMBER2019.indd 16 10/25/19 3:25 PM columns
social media was used to pro- Identifying Climate mote the event. Altogether, the Change Vulnerabilities C40 hackathon was both a great in the Third World meeting of minds inspiring cre- Whether due to deforestation for ative solutions and a roadmap for farmland, pollution, overpopu- future events. lation, or other mechanisms, harmful practices in poor coun- Protecting Endangered tries are often contributors to Species with Machine climate change. Helping to meet Learning this need, the Global Facility for Another area in which Data for Disaster Reduction and Recovery Good is making a difference (GFDRR, www.gfdrr.org/en) col- in our natural world is animal laborated with Data for Good conservation. Monitoring the leader DataKind. movement of animals provides The GFDRR is a World critical data for conservations of Bank agency providing funding threatened and endangered spe- for projects addressing the risk Getting Involved cies. However, the devices used for of disasters around the world. A In Data for Good opportunities this month, tracking can be expensive, difficult UK-based DataKind team of vol- check out the student awards at http://asa. to implement, and short lived. unteer data scientists developed stat.uconn.edu. Sponsored by the Statistical They can also present a risk of a machine learning algorithm to Computing and Statistical Graphics sections, interfering with animal behavior, analyze satellite images, identify- possibly altering the very qualities ing buildings and their type and two student awards will be presented at JSM in researchers want to measure. use (www.elsevier.com/connect/ Philadelphia. The conservation group new-in-disaster-science-using- The John M. Chambers Statistical Software WildTrack (https://wildtrack.org) machine-learning-and-maps-to- Award honors software written by, or in has developed a system to iden- see-whos-vulnerable). collaboration with, an undergraduate tify individuals from several spe- This information can be used or graduate student. The Student Paper cies of animals from footprints. following a natural disaster to Photographs of the footprints identify population centers, criti- Competition recognizes a paper in the area of are digitized, and a machine cal infrastructure, and surviving statistical computing and statistical graphics. learning algorithm identifies the facilities to help aid agencies Neither are specific to Data for Good, but it’s a animal that made it with 90 per- respond more quickly and effec- great opportunity to show off your work while cent accuracy. tively. This analysis makes the making a positive difference with statistics. I hope These algorithms are being countries using them more resil- the D4G community will be well represented. developed by more than a dozen ient to natural disasters. university students around the More projects are in the works, The submission deadline for both awards is world. To date, there are algo- connecting volunteer statisticians December 15. rithms for 13 species. and data scientists to address UN Also of interest to Data for Good advocates and JMP, from the SAS Institute, sustainable development goals. supports the project and provides Data for Good opportuni- activists is the new ASA Leader HUB (https:// its data visualization software. ties to help preserve and protect community.amstat.org/asaleaderhub/home). The result of this collaboration the world around us are as varied It’s a central location for resources leaders in between an NGO, universities, as the concerns that need to be ASA chapters, sections, and interest groups can and industry is an inexpensive, addressed. When we participate use. Of particular interest is the ASA Chapter noninvasive, and effective method in Data for Good, each and every Stimulus program, supporting activities that of counting and monitoring threat- one of us has the power to become develop chapter interest and membership— ened and endangered animals. a tipping point—a point of lever- and Data for Good projects are a great way to age to bring about beneficial do it. change and create data-driven sus- tainable practices for better com- munities and a healthier world. n
november 2019 amstat news 17 JSM2020Flyer.pdf 1 6/20/19 4:48 PM
E DATES FOR PARTICIPANTS
July 18, 2019 – September 5, 2019 Invited Session Proposal Submission
September 30, 2019 Continuing Education Course Proposal Deadline
November 12, 2019 – December 10, 2019 Topic-Contributed Session Proposal Submission
December 3, 2019 – February , 2020 General Abstract Submission (all except invited papers and panels)
January 15, 2020 Computer Technology Workshop Proposal Deadline
January 22, 2020 – April 2, 2020 Meeting and Event Request Submission
February 3, 2020 – April 15, 2020 Late-Breaking Session Proposal Deadline
May 15, 2020 Draft Manuscript Deadline