Subscribe to AmstatNews feed
Updated: 41 min 43 sec ago

JSM Is Baltimore Bound!

Mon, 05/01/2017 - 7:00am

With more than 3,000 individual presentations arranged into approximately 204 invited sessions, 300 contributed sessions, and 500 poster and speed poster presentations, the 2017 Joint Statistical Meetings will be one of the largest statistical events in the world. It will also be one of the broadest, with topics ranging from statistical applications in numerous industries to new developments in statistical methodologies and theory. Additionally, there will be presentations about some of the newer and expanding boundaries of statistics, such as analytics and data science.

More About JSM

This year, the exhibit hall is the place to be. The Opening Mixer will take place there in addition to Spotlight Baltimore, which will feature events throughout the week. Moreover, if you are looking for a way to help the local community, you’ll want to visit IMPACT Baltimore. Finally, there will be an art show featuring data artists just inside the hall.

Here are the featured speakers you can expect at JSM 2017. We hope to see you there.


Featured Speakers Monday, July 31

8:30 a.m.
IMS Medallion Lecture I
Edoardo M. Airoldi, Harvard University

“Design and Analysis of Randomized Experiments on Networks”

Classical approaches to causal inference largely rely on the assumption of “no interference,” according to which the outcome of an individual does not depend on the treatment assigned to others. In many applications, however, such as evaluating the effectiveness of health care interventions that leverage social structure or assessing the impact of product innovations on social media platforms, assuming lack of interference is untenable. In fact, the effect of interference itself is often an inferential target of interest, rather than a nuisance. In this lecture, we will formalize technical issues that arise in estimating causal effects when interference can be attributed to a network among the units of analysis, within the potential outcomes framework. We will then introduce and discuss several strategies for experimental design in this context.

10:30 a.m.
Blackwell Lecture
Martin Wainwright, University of California, Berkeley

“Information-Theoretic Methods in Statistics: From Privacy to Optimization”

Blackwell made seminal contributions to information theory and statistics, including early work on characterizing the capacities of various channels. The notion of channel capacity has a natural analogue for statistical problems, where it underlies the characterization of minimax rates of estimation. Inspired by this seminal work, this talk is devoted to the use of information-theoretic methods for tackling statistical questions, and I provide two vignettes. First, in the realm of privacy-aware statistics, how to characterize the tradeoffs between preserving privacy and retaining statistical utility? Second, in the realm of statistical optimization, how can we characterize the fundamental limits of dimensionality reduction methods?

This lecture will draw on joint work with John Duchi, Michael Jordan, and Mert Pilanci.

2:00 p.m.
IMS Medallion Lecture II
Emery N. Brown, Massachusetts Institute of Technology

“State-Space Modeling of Dynamics Processes in Neuroscience”

Dynamic processes are the rule, rather than the exception, in all areas of neuroscience. For this reason, many of the data analysis challenges in neuroscience lend themselves readily to formulation and study using the state-space paradigm. In this lecture, I will discuss the state-space paradigm using both point process and continuous valued observation models in the study of three problems in basic and clinical neuroscience research: characterizing how the rodent hippocampus maintains a dynamic representation of the animal’s position in its environment; real-time tracking of brain states of patients receiving general anesthesia; and real-time assessment and control of medical coma. This research has led to the development of state-space methods for point process observation models and state-space multitaper methods for time-frequency analysis of nonstationary time series.

4:00 p.m.
ASA President’s Invited Address
Jo Craven McGinty, The Wall Street Journal

Abstract unavailable.


8:00 p.m.
IMS Presidential Address and Awards Ceremony
Jon A. Wellner, University of Washington
“The IMS at 82: Past, Present, and Future”

In 2017, the IMS reaches the age of 82. Will it make it to 100? In this talk, I will argue that the IMS has succeeded remarkably well in fulfilling its fundamental goal: “To foster the development and dissemination of the theory of statistics and probability.” The IMS publishes the top journals in both statistics and probability and organizes world-class meetings on a regular basis in conjunction with its sister/brother organizations, the Bernoulli Society and the ASA.

As noted by Jim Pitman in his 2008 IMS Bulletin article, “… [A]mong many organizations with this goal, the IMS stands out as the most responsive to creative suggestions about how to achieve it.” In this talk, I will review the history of the IMS, summarize the current state of affairs of the IMS as an organization and its goals, and briefly discuss future directions. With continued creative responsiveness to recent trends, it seems likely the IMS will easily reach its 100th anniversary.


Tuesday, August 1

8:30 a.m.
IMS Medallion Lecture III
Subhashis Ghoshal, North Carolina State University

“Coverage of Nonparametric Credible Sets”

The celebrated Bernstein-Von Mises theorem implies that for regular parametric problems, Bayesian credible sets are also approximately frequentist confidence sets. Thus, the uncertainty quantification by the two approaches essentially agree, even though they have very different interpretations. A frequentist can then construct confidence sets by Bayesian means, which are often easily obtained from posterior sampling. However, the incredible agreement can fall apart in nonparametric problems whenever the bias becomes prominent. Recently, positive results have appeared in the literature overcoming the problem by undersmoothing or inflation of credible sets. We shall discuss results on Bayes-frequentist agreement of uncertainty quantification in white noise models, nonparametric regression, and high-dimensional linear models. We shall also discuss related results on nonlinear functionals.

2:00 p.m.
IMS Medallion Lecture IV
Mark Girolami, Imperial College London

“Probabilistic Numerical Computation: A Role for Statisticians in Numerical Analysis?”

A research frontier has emerged in scientific computation founded on the principle that numerical error in numerical methods that, for example, solve differential equations entails uncertainty that ought to be subjected to statistical analysis. This viewpoint raises exciting challenges for contemporary statistical and numerical analysis, including the design of statistical methods that enable the coherent propagation of probability measures through a computational and inferential pipeline. A probabilistic numerical method is equipped with a full distribution over its output, providing a calibrated assessment of uncertainty shown to be statistically valid at finite computational levels, as well as in asymptotic regimes. The area of probabilistic numerical computation defines a nexus of ideas, philosophies, theories, and methodologies bringing together statistical science, applied mathematics, engineering, and computing science. This talk seeks to make a case for the importance of this viewpoint. I will examine the case for probabilistic numerical methods in mathematical modeling and statistical computation while presenting case studies.

4:00 p.m.
ASA Deming Lecture
Fritz Scheuren, NORC at the University of Chicago

“A Rake’s Progress Revisited”

Deming’s statistical consulting advice to us is rich and varied. I knew Deming a little and long loved him from afar. That affection for this great man should be evident in this talk. To give focus to my remarks, I will cover just one of his quality ideas in depth: the algorithm commonly called “raking.”

Raking, or raking ratio estimation, was advanced by Deming and Stephan for use in the 1940 U.S. Decennial Census. The algorithm employs at its heart a process whereby the weights of a data set are iteratively ratioed-adjusted within categories, until they simultaneously meet, within tolerance, a set of pre-specified population totals.

Some confusion surrounding raking’s introduction slowed its development. At its base, the approach was intuitive. Its theoretical development came long after, and some of the justifications for its use were misplaced, including by Deming.

For a long time, the lack of computing power limited raking to modest applications and, given the work needed, the benefit/cost ratio often seemed insufficient relative to the time and money that had to be expended. These limitations no longer hold.

The problem of variance calculation accounting for raking posed additional challenges. These were solvable asymptotically in some decennial and sample survey settings, but usually not in a closed form. Even now, replication techniques are most commonly the only practical solution available for general sample survey settings.

This talk will motivate these assertions with examples taken from my practice and that of other statisticians. Extensions by me to multivariate raking will also be covered and speculations on unsolved or incompletely posed problems will be offered. Throughout, I will intersperse examples from my decades of practice.

8:00 p.m.
ASA President’s Address and Founders & Fellows Recognition
Barry D. Nussbaum

“Statistics: Essential Now More Than Ever (Or, Why Uber Should Be in the Driver’s Seat for Cars, Not for Data Analysis)”

Now is the time for statisticians to come to the rescue of rational analysis of today’s challenges. Our profession is essential now, perhaps more than ever before. The world is besieged with a deluge of Big Data and analysts ready to process it. Rather than being captive to this burgeoning field, statisticians are poised to provide the critical elements for affecting social problems. This talk focuses on the new initiatives taken to ensure the proper and vital use of statistics now and into the future. Let the Ubers of the world be in the driver’s seat to speed you to the airport, while the statisticians correctly assess the data. Examples will be given showing the crucial importance and vital strength of statisticians who are at the table when decisions are made. The art of effective collaboration with clear and succinct explanations is more important than ever. Come hear how you can improve your contributions to society and our collective standing in the world.


Wednesday, August 2

8:30 a.m.
IMS Medallion Lecture V
Judith N. Rousseau, Université Paris Dauphine

“On the Semiparametric Bernstein-Von Mises Theorem in Regular and Nonregular Models”

In regular models, the renown Bernstein-Von Mises theorem states that the posterior distribution of a quantity of interest, say $\theta$, is asymptotically Gaussian with mean $\hat \theta$ and variance $V/n$ when the data are assumed to be distributed from a model $P_{0}$. It also states that, under $P_0$, $\sqrt{n}( \hat \theta- \theta_0) $ is asymptotically Gaussian with mean zero and variance $V$. This duality between the asymptotic behavior of the posterior distribution of $\theta$ and the frequentist distribution of $\hat \theta$ has important implications in terms of strong adequacy between the Bayesian and frequentist approaches. In non-regular models, a similar adequacy can happen; however, the asymptotic distribution may not be Gaussian nor the concentration rate by $1/\sqrt{n}$. These results are well known in parametric models.

In this talk, I will present developments that have been obtained in both regular and non-regular semiparametric models (i.e., when the parameter of interest $\theta$ is finite dimensional, but the model also includes an infinite or high-dimensional nuisance parameter).

4:00 p.m.
COPSS Awards and Fisher Lecture
Robert E. Kass, Carnegie Mellon University

“The Importance of Statistics: Lessons from the Brain Sciences”

The brain’s complexity is daunting, but much has been learned about its structure and function, and it continues to fascinate. On the one hand, we are all aware that our brains define us; on the other hand, it is appealing to regard the brain as an information processor, which opens avenues of computational investigation.

While statistical models have played major roles in conceptualizing brain function for more than 50 years, statistical thinking in the analysis of neural data has developed much more slowly. This seems ironic, especially because computational neuroscientists can, and often do, apply sophisticated data analytic methods to attack novel problems. The difficulty is that, in many situations, trained statisticians proceed differently than those without formal training in statistics. What makes the statistical approach different and important? I will give you my answer to this question and go on to discuss a major statistical challenge, one that could absorb dozens of research-level statisticians in the years to come.


Wald Lectures

Emmanuel J. Candes, Stanford University
“What’s Happening in Selective Inference?”
Wald Lecture I
4:00 p.m., Tuesday, August 1
Wald Lecture II
10:30 a.m., Wednesday, August 2
Wald Lecture III
8:30 a.m., Thursday, August 3

Science has long operated as follows: A scientific theory can only be empirically tested, and only after it has been advanced. Predictions are deduced from the theory and compared with the results of experiments so they can be falsified or corroborated. This principle, formulated by Popper and operationalized by Fisher, has guided the development of scientific research and statistics for nearly a century. We have, however, entered a new world where large data sets are available prior to the formulation of scientific theories. Researchers mine these data relentlessly in search of new discoveries, and it has been observed that we have run into the problem of irreproducibility. Consider the April 23, 2013, Nature editorial: “[…] Nature has published a string of articles that highlight failures in the reliability and reproducibility of published research.” The field of statistics needs to reinvent itself to adapt to the new reality in which scientific hypotheses/theories are generated by data snooping. I will make the case that statistical science is taking on this great challenge and discuss exciting achievements such as FDR theory, knockoffs, and post-selection inference.


Diversity Mentoring Program Accepting Applications

Mon, 05/01/2017 - 7:00am

Diversity Mentoring participants at JSM 2016

Applications are being accepted for the JSM Diversity Mentoring Program, which brings minority graduate and undergraduate students, postdoctoral scholars, and junior professionals together with senior-level statisticians and faculty in academia, government, and the private sector in a structured program during the Joint Statistical Meetings. The program provides career information, mentoring, and a peer network. Program activities include small-group discussions and one-on-one meetings between mentor-protégé pairs.

Apply online. Preference will be given to applications received by May 31. For more information, contact Dionne Swift.

This program is sponsored by the American Statistical Association and the ASA Committee on Minorities in Statistics.

JSM Student Opportunities

Mon, 05/01/2017 - 7:00am

Lara Harmon, ASA Community Coordinator

Students, will you be joining us at JSM 2017? We hope to see you there! We’re in Baltimore this year, and we’ve got plenty of opportunities lined up for you to get involved and learn more about what the ASA offers.

Would you like to …

Attend a Continuing Education (CE) course for free?

JSM’s CE courses give conference attendees a chance to dive into a new topic, brush up skills they already possess, and benefit from the knowledge of experts they might never otherwise have the chance to learn from in person. You can take advantage of all these benefits for free! As a CE monitor, you will help the course run smoothly while also observing course content and meeting course attendees and instructors. Interested? Contact Rick Peterson, the ASA’s professional development and chapters and sections manager. Keep an eye on the JSM 2017 website for a list of CE courses.

Meet presenters and get experience managing a session?

Session chairs introduce and support session presenters. The work of these volunteers helps keeps sessions on topic and running on time. It also gives the chairs the opportunity to meet and work directly with session presenters. Our session chair slots are full for JSM 2017, but you can volunteer to be an emergency backup. To learn more about what session chairs do, visit the JSM 2017 website, scroll down to the “Chairs” bar, and click to see the details.

Learn more about student chapters?

The ASA established a student chapter program several years ago, and it’s taken off—we now have more than 50 student chapters, thanks to your energy and enthusiasm! This year, we’re planning our first JSM student chapter event. Join us to learn from the experiences of a panel of student chapter presidents and faculty advisers, and then break out into small groups to share your own ideas as we brainstorm about the future of student chapters. Time and place TBA. Contact Lara Harmon, marketing and online community coordinator if you’re interested in helping with setting up and moderating the event.

Meet fellow student members from across the country (and the world)?

If you’ve been to JSM before, you’ve probably experienced the JSM student mixer. It’s not to be missed! Join students from across the U.S. and abroad, both first-time attendees and seasoned pros, as we gather for drinks, snacks, and a free raffle on July 31 from 6–8 p.m. Prizes in past years have ranged from stats software packages to an Xbox 360 (so you might want to leave a little room in your carry-on luggage if you plan on attending)! We always appreciate a little extra help handing out raffle tickets before the mixer; if you’re interested, contact Lara Harmon.

Relax and put your feet up while helping children in need?

Networking and walking from session to session (to session to session) can wear you out, mentally and physically. If you need a break, check your program for IMPACT Baltimore activities. Head over to the exhibit hall at the times shown in the program and you can help make no-sew blankets for children in need. Last year, ASA members used up all the blanket-making materials we brought, and we’re hoping to repeat our success this year. The blankets are easy to make and all supplies are provided, so grab some friends and do your daily good deed.


Deviate from the standard conference routine and get to know your fellow ASA members in a casual environment. After meeting up with your peers at the Monday student mixer, we give you the chance to dance on Tuesday! A JSM tradition, the dance party invites you to mingle and chat or get out on the floor. Either way, you can relax, catch up on others’ JSM experiences, and share your plans for the rest of the conference.

We hope you can come!

What’s in Baltimore Besides JSM? More Than You Think!

Mon, 05/01/2017 - 7:00am

The Baltimore Convention Center is located at 1 W. Pratt St., just two blocks from Baltimore’s famous Inner Harbor. The location makes it convenient to squeeze a little “tourist time” into your busy JSM schedule. Baltimore’s Inner Harbor is home to a number of interesting attractions, including the following:

National Aquarium
501 E. Pratt St.
Open daily, see website for hours and ticket prices
Strongly recommend purchasing tickets in advance

The National Aquarium in Baltimore houses sharks, dolphins, rays, and tropical fish among the more than 17,000 creatures in naturalistic exhibits, including a walk-through rain forest, a 4-D immersion theater, the Living Seashore touch pool, and an Australian exhibit featuring a 35-foot waterfall.

Maryland Science Center
601 Light St.
Open daily, check website for hours and prices
Strongly recommend purchasing tickets in advance

Explore numerous hands-on activities at the Maryland Science Center. Featuring an IMAX theater and planetarium, it’s sure to please all visitors. Don’t miss the new exhibit, Science & Main, that takes you past Baltimore landmarks as you learn how science interacts with your everyday life.

Historic Ships
Pier 1, 301 E. Pratt St.
Sunday–Thursday, 10:00 a.m. – 5:00 p.m.; Friday, 10:00 a.m. – 6:00 p.m.; Saturday, 10:00 a.m. – 7:00 p.m.
See website for ticket prices

Be sure to witness the historic ships in Baltimore by touring the USS Constellation, the USS Torsk, the USCGC Taney, and the Lightship Chesapeake. Learn about the role these vessels played during various battles in American history. Also tour Seven Foot Knoll Lighthouse.

Cheer on the BIRDS!
Baltimoreans take their sports seriously. Today, sports fans flock to Oriole Park at Camden Yards, the first of the new breed of retro ballparks, to cheer on the O’s—or the Birds as they are known locally. Take this opportunity to watch the O’s host the Kansas City Royals July 31–August 2 or take on the Tigers August 3–6. Can’t make a game? Tours are also available, game schedules permitting.

But wait, there is more! Just a bit farther, but easily accessible with public transportation, are the following:

Baltimore Museum of Art
10 Art Museum Drive
Wednesday – Sunday, 10:00 a.m. – 5:00 p.m.
Entry to the BMA is free for everyone

Home to an internationally renowned collection of 19th-century, modern, and contemporary art, the BMA includes one of the most important African collections in the country. Founded in 1914 with a single painting, the BMA today has 95,000 works of art—including the largest holding of works by Henri Matisse in the world. The museum has a long tradition of collecting the art of the day, beginning with the Cone Sisters, whose acquisitions from living artists led to the museum’s commitment to contemporary art.

B&O Railroad Museum
901 W. Pratt St.
Open Monday – Saturday, 10:00 a.m. – 4:00 p.m.; Sunday, 11:00 a.m. – 4:00 p.m.
Entrance fee: Adults $18; Seniors (60+) $16; Children (2–12) $12

This fascinating and fun place for kids, families, and lovers of history features the most important railroad collection in America, as well as seasonal train rides and free parking.

Walters Art Museum
600 N. Charles St.
Wednesday – Sunday, 10:00 a.m. – 5:00 p.m.; Thursday, 10:00 a.m. – 9:00 p.m.

The collection at the Walters Art Museum presents an overview of world art from pre-dynastic Egypt to 20th-century Europe and counts among its many treasures Greek sculpture, Roman sarcophagi, medieval ivories, Old Master paintings, Art Nouveau jewelry, and 19th-century European and American masterpieces. Try out the new mobile guide to discover the stories behind the collection.

Fort McHenry National Monument and Historic Shrine
2400 E. Fort Ave.
Open daily, 9:00 a.m. – 5:00 p.m.
Entrance fee: $10 for adults; children 15 years and younger are free

At Fort McHenry, you can learn about the Battle of Baltimore and the birth of the “Star Spangled Banner” and experience events like living history weekends, where the Fort McHenry Guard performs demonstrations—all just a water taxi ride away from the Inner Harbor.

Getting Around

Charm City Circulator
Be sure to take advantage of Baltimore’s Charm City Circulator, a fleet of free bus shuttles that travel four routes throughout the city. Operating every day, check out the Circulator website to download the Circulator app to use while moving around the city.

Monday – Thursday: 7:00 a.m. – 8:00 p.m.
Friday: 7:00 a.m. – midnight
Saturday: 9:00 a.m. – midnight
Sunday: 9:00 a.m. – 8:00 p.m.

Baltimore Water Taxi
This water transportation system of 17 blue-and-white boats is the oldest of its kind in the country. Offering one price for all-day, unlimited on-off service to more than 30 attractions and neighborhoods—including Fell’s Point, Canton Waterfront Park, and Fort McHenry—don’t miss this fun way to get around Baltimore.

Statistical Education Section News for May

Mon, 05/01/2017 - 7:00am
Dalene Stangl and Kelly McConville

The 2017 JSM program will include nearly 100 speakers sponsored by the Statistical Education Section. These speakers will appear throughout three invited sessions, five topic-contributed sessions, five contributed sessions, 24 speed/posters, and 11 roundtables.

Invited Sessions
  • Modernizing the Undergraduate Statistics Curriculum
    Speakers: Nick Horton, Hilary Parker, Jo Hardin, and Colin Rundel
  • Novel Approaches to First Statistics / Data Science Course
    Speakers: Ben Baumer, Mine Çetinkaya-Rundel, Rebecca Nugent, and Daniel Kaplan
  • Training Statisticians to Be Effective Instructors

    Panelists: Adam Loy, Jennifer Kaplan, Meghan Short, Patricia Buchanan, and Paul Stephenson

Topic-Contributed Sessions
  • Being Research Active in Teaching-Focused Colleges
  • The Essential Connections Between Industry and Statistics Education: Innovation, Technology, and Partnerships
  • Design, Implementation, and Impact of Different Approaches to Professional Development for Teachers of Statistics
  • Teaching Introductory Statistics Using Simulation-Based Inference Methods
  • Modernizing the Statistical Collaboration Course
Contributed Sessions
  • Advances in Pedagogy
  • Technologies in the Classroom
  • Teaching Special Groups and Underate Researi>
  • Teaching Introductory Statistics iostatistics
  • Topics in Math/Stat andourses

We have a great slate of roundtables this year. Roundtables are an informal, themed discussion over breakfast or lunch. They are a great way to meet educators from other institutions. Registration for roundtables opens with general registration on May 1.

A.M. Roundtables

  • Infusing Data Science into the Statistics Curriculum
  • Turning a Tweet into a Lesson: Using Current Events as a Context
  • Introducing Bayesian Statistics at Courses of Various Levels
  • Why Do Students Hate Statistics?

P.M. Roundtables

  • A Course in Business Analytics
  • Student Involvement in Community Projects
  • Discussing the Uses and Creation of R Shiny Applications
  • Incorporating Complex Survey Concepts into the Curriculum
  • Recruiting and Engaging Students
  • GAISEing at a Lecture Hall: Effective Pedagogy in Large-Enrollment Courses
  • What Are the 25 Most Common Terms in Statistics from the Last 20 Years?

For more information about the JSM program, view the online program.

Is This Your First JSM? Here Are Tips to Navigate the Conference

Mon, 05/01/2017 - 7:00am

Christopher Bilder is a professor in the department of statistics at the University of Nebraska-Lincoln. He will be presenting the continuing education course “Analysis of Categorical Data” during JSM. He earned his PhD in statistics from Kansas State University.

The largest congregation of statisticians in the world happens every August during the Joint Statistical Meetings (JSM). More than 6,000 people attend these meetings, which are sponsored by 11 statistical societies, including the American Statistical Association. The meetings offer a variety of activities such as attending research presentations, interviewing for jobs, taking professional development courses and workshops, and browsing the exhibit hall. With so many opportunities, new attendees can be overwhelmed easily by their first JSM experience.

Based on my familiarity with attending meetings over the last 16 years and the experiences of student groups I have led, I’m going to tell you how to get the most out of JSM. If you would like to share your own recommendations, I encourage you to submit a comment.

Most new attendees who choose to present their research do so in a contributed session via an oral or poster presentation. The deadline to submit an abstract for acceptance into the program was in early February. For those who did this, additional proof of progress (e.g., drafts of a paper) for the presentation must be submitted by mid-May.

Before JSM

A preliminary program listing the presentation schedule is now available. Because there may be more than 40 concurrent presentations at any time, it is best to arrive at JSM with an idea of which to attend. This can be done by examining the session titles and performing keyword searches in the online program prior to JSM.

Oral presentations are separated into invited, topic-contributed, and contributed sessions, with each session lasting 1 hour and 50 minutes. Invited and topic-contributed sessions include groups of related presentations that were submitted together and selected by JSM Program Committee members. These presentations each last for 25 or more minutes for invited and 20 minutes for topic-contributed. Contributed paper sessions include groups of 15-minute oral presentations. Unlike invited and topic-contributed sessions, contributed presentations are submitted individually and then grouped by JSM Program Committee members.

Poster presentations are also separated into invited, topic-contributed, and contributed sessions, with the vast majority in contributed sessions. These types of presentations involve speakers being available for questions next to their displayed poster during the entire session. Most posters are of the traditional paper format, but an increasing number now are in an electronic format. This latter format involves a large, high-definition TV that shows the poster all at once or cycles through a small number of slides that would normally be printed on paper. Relatively new to JSM is a hybrid of an oral and poster presentation. The oral poster presentation component begins with a “speed session,” in which four-minute presentations are given by each speaker. Later the same day, electronic posters are made available for these same presentations.

Online registration for JSM begins around May 1. For members of a sponsoring statistical society, the cost is $435 during the early registration period. The cost increases to $535 if you register at JSM. Registration for student members is only $110, and this rate is available at any time. Also starting around May 1, you can reserve a hotel room through the JSM website. A number of hotels near the convention center are designated as official conference hotels, and they discount their normal rates. However, even with a discount, you can expect to pay $200 or more per night for a room.

Attending JSM can be expensive. Students have several options to reduce the cost burden. First, ask your adviser or department for funding. Many departments offer financial support for students who present their research at JSM. Students also may qualify for funding from the student activities office on their campus. For example, when I was a student, my department’s statistics club received funding this way, which paid for most of my first JSM expenses.

In addition to school-based resources, many ASA sections sponsor student paper competitions that provide travel support to award winners. For example, the Biometrics Section of the ASA sponsors the David P. Byar Young Investigators Award, with $2,000 awarded to the winner and separate $1,000 awards given to authors of other outstanding papers. Most competitions require a completed paper to be submitted many months prior to JSM.


JSM begins on a Sunday afternoon in late July. Business casual clothing is the most prevalent attire, but some attendees wear suits and others wear T-shirts and shorts. When you arrive at JSM, go to the registration counter at the convention center to obtain your name badge and additional conference materials.

There is a significant online presence during JSM. A main resource is the JSM app and online program. Both contain all the information you’ll need and more. Also, the ASA posts the most up-to-date news about JSM through its Twitter (@AmstatNews) and Facebook accounts. Attendees at JSM can use #JSM2017 to tag their JSM-related posts.

To welcome and orient new attendees, the JSM First-Time Attendee Orientation and Reception is scheduled for early Sunday afternoon. At this reception, docents will be present (identified with a special orange button by their name badge) to answer any questions you may have about the meetings. These docents will be available throughout the conference as well.

Later on Sunday evening, the Opening Mixer will be held in the exhibit hall. This event is open to all attendees, and drinks and hors d’oeuvres will be served.

In between the orientation and the mixer, the ASA Awards Celebration and Editor Appreciation session is held. Many first-time attendees are honored during it due to being awarded a scholarship or winning a student-paper competition.

The main sessions start Sunday at 2:00 p.m. Many of the research presentations are difficult to understand completely. My goal for a session is to have 1–2 presentations in which I learn something relevant to my teaching or research interests. This may seem rather low, but these items add up after attending many sessions.

For attendees who teach introductory courses, the sessions sponsored by the ASA Section on Statistical Education are often the easiest to understand. Many of these sessions share innovative ideas about how to teach particular topics.

Introductory overview lectures are another type of session that has easier-to-understand topics. Recent lectures have included introductions to Big Data, bioinformatics, and complex survey sampling. There are also many Professional Development courses and workshops available for an additional fee. However, you can attend a course for free by volunteering prior to JSM to be a monitor. Monitors perform duties such as distributing and picking up materials during the course. As an added benefit, monitors can attend one additional course for free without any duties. Those who are interested should contact Rick Peterson.

Featured talks at JSM are usually scheduled for late afternoon on Monday through Wednesday. On Tuesday evening, the ASA presidential address is given, along with an introduction to the new ASA fellows and winners of the Founders Award. The fellows introduction is especially interesting because approximately 50 ASA members (<0.33% of all members) are recognized for their contributions to the statistics profession.

In addition to presentations, the JSM exhibit hall features more than 70 companies and organizations exhibiting their products and services. Many exhibitors give away free items (e.g., candy, pens, etc.). All the major statistics textbook publishers and software companies are there. Textbook publishers usually offer a discount on their books during JSM and often for a short time after. The exhibit hall also includes electronic charging stations and tables that can be used for meetings. It’s also the location for the poster presentations.

The JSM Career Service provides a way for job seekers and employers to meet. Pre-registration is required, and the fee is discounted if you register before mid-July. The service works by providing an online message center for job seekers and employers to indicate their interest in each other. Once a common interest is established, an interview can be arranged for during the meetings.

Other activities at JSM include the following:

  • Shopping at the ASA Store to purchase a statistics-themed T-shirt or mug
  • Attending an organized roundtable discussion during breakfast or lunch about a topic of interest (pre-registration is required)
  • Taking a little time off from JSM for sightseeing or attending a sporting event
After JSM

JSM ends in the early afternoon on Thursday. Don’t let what happens at JSM stay at JSM! The first thing I do after the meetings is prepare a short review of my activities. Using notes I took during sessions, I summarize items from presentations I want to examine further.

I also summarize meetings I had with individuals about research or other important topics. Much of this review process starts at the airport while waiting for my return flight.

If you give a presentation at JSM, you may submit a corresponding paper to be published in the conference proceedings. Papers are not peer-reviewed in the same manner as for journals, but authors are encouraged to have others examine their paper before submission. The proceedings are published online around December. Authors retain the right to publish their research later in a peer-reviewed journal.

AMS Announces Bertrand Russell Prize

Mon, 05/01/2017 - 7:00am

The American Mathematical Society (AMS) has a new prize: Bertrand Russell Prize of the AMS. The prize was established by Thomas Hales of the University of Pittsburgh and honors research or service contributions of mathematicians or related professionals to promoting good in the world. It also recognizes the ways mathematics furthers human values.

Nominations are being accepted until June 30 for the 2018 prize.

Interview with Editor of Statistics in Biopharmaceutical Research

Mon, 05/01/2017 - 7:00am

Frank Bretz is the editor of Statistics in Biopharmaceutical Research. We asked him to tell us a little about himself, the journal, and what we can expect to read in future issues.

Where did you grow up and go to school, and what or who inspired you to be a statistician?

I grew up on three continents, until I settled in Germany to study mathematics with a minor in biology. It was not until my PhD, however, that I developed a passion for statistics and recognized the necessity to bridge the gap between methodological developments and their applications in the day-to-day world.

Why did you become interested in being the editor for SBR?

The role of statistics in biopharmaceutical research is ever increasing, and SBR has played a major role in this regard since its launch in 2009. Thus, it is an honor to build upon the work of the previous editors and follow the call from Joe Heyse in the journal’s first issue to “publish original peer-reviewed articles directed to researchers and applied statisticians from academia, government, and industry supporting the growing disciplines in the biopharmaceutical sciences.”

Do you plan to make any changes to the journal while you are editor?

We need to acknowledge the globalization of the pharmaceutical industry. Complementing my other services in supporting the ASA’s international outreach activities—particularly for the Biopharmaceutical Section—I would like to increase the visibility of SBR outside the United States. As a first step, we have enriched the editorial board with selected members from other regions to broaden the impact of SBR and encourage submissions from outside the United States.

At the same time, I would like to sharpen SBR’s profile at the scientific interface between industry, academia, and regulatory agencies to advance practices of pharmaceutical drug development. This could be done by, for example, establishing dedicated sections such as a “regulatory corner” for a scientific exchange on newly developed statistical guidelines or a section on “out of box” statistical methods and practices.

What do you find most enjoyable about being a journal editor?

For me, I found that my past experiences as associate or guest editor for various journals was extremely rewarding. I found I was involved in handling new science at a detailed level unmatched by other venues. I also found myself getting new contacts in the form of authors and reviewers.

When the opportunity came around for me to take on the editorial responsibility for SBR, I did not need much convincing, but the reasons were more personal than anything else. Becoming an editor is not something you just choose; you need to see if you personally get something out of the job that makes it worth your while.

Ultimately, the most enjoyable part is the fact that I’m surrounded by an outstanding team that shoulders much of the heavy work: Jina Lee as the editorial coordinator, Eric Sampson as the journal manager, and, last but not least, the entire editorial board.

What do you find is the most challenging part of your job as editor?

To sign up as chief editor likely means signing up for a longer period to be able to fully embrace the flow of articles and handle all the problems that may occur. This is particularly true if you wish to see any results of your work while you are still associated with the journal.

What do you enjoy doing in your spare time?

Traveling, hiking, reading, immersing into different cultures.

Health Policy Statistics Section Achievement Awards Open

Mon, 05/01/2017 - 7:00am

The 2018 HPSS Achievement Awards honor individuals who have made significant contributions to the development of statistical methods or have developed innovative statistical applications for health care policy or health services research to encourage research in this area and to increase awareness of the HPSS.

The HPSS Mid-Career Award recognizes someone who has shown leadership in the field of health care policy and health services research through outstanding contributions of methodological or applied work and the promise of continued excellence at the frontier of statistical practice that advances the aims of HPSS.

Candidates must be within 15 years of their terminal degree on January 1, 2017, and cannot be a previous HPSS Mid-Career Award winner. The 2015 winner is Elizabeth A. Stuart of The Johns Hopkins University.

The HPSS Long-Term Excellence Award recognizes significant contributions to health care policy and health services research through mentoring and/or service that advance the aims of HPSS.

Candidates cannot be within 15 years of their terminal degree on January 1, 2017, and cannot be a previous HPSS Long-Term Excellence Award winner. The 2015 winners are Constantine A. Gatsonis of Brown University and Donald Hedeker of The University of Chicago.

These awards will be presented at the 12th International Conference on Health Policy Statistics (ICHPS), January 10–12, 2018, in Charleston, South Carolina.

Nominations should include the following:

  • The nominee’s curriculum vitae
  • A letter (not to exceed two pages) from the nominator summarizing the nominee’s credentials
  • Additional independent (other than the nominator or nominee) evaluation letters (not required)
  • Contact information for the nominee or nominator (if different)

Nominations should be sent by midnight on August 11 to the awards committee. Questions also should be directed to the committee.

Survey Research Methods Section News for May

Mon, 05/01/2017 - 7:00am

The Online Proceedings of the Survey Research Methods Section (SRMS) for the 2016 Joint Statistical Meetings, held in Chicago, are available online now.

The section is actively searching for a student to work on assembling the proceedings for the next JSM (with a $500 stipend). If you are a student interested in working on this project, contact the section’s publication officer, Tony An.

2017 JSM Updates

JSM 2017, to be held in Baltimore, is fast approaching. The Survey Research Methods Section sponsors 10 invited sessions, 13 topic-contributed sessions, and 12 contributed sessions. In addition, the section sponsors traditional and speed poster sessions, three full-day short courses, and two P.M. roundtables. Here is a preview of the 2017 lineup:

Continuing Education Courses

  • Synthetic Data Sets for Statistical Disclosure Limitation

    Led by Jörg Drechsler of the Institute for Employment Research, IAB, Nuremberg, Germany, and Jerry Reiter of Duke University

    This course focuses on practical aspects of confidentiality protection and provides an overview of modeling strategies, analytical validity evaluations, and potential measures to quantify the remaining risk of disclosure, with illustrations of R programming.

  • Construction of Weights in Surveys

    Led by David Haziza of the University of Montréal

    This course offers a detailed description of weighting methods, including inversion of probability of selection, nonresponse adjustment, calibration, and trimming adjustments.

  • Research and Analysis Workflows: Low-Cost, Every-Day Project Management Techniques, Tools, and Tips That Produce High-Quality, Streamlined, Stress-Free Research and Data Science

    Led by Matt Jans of Abt Associates and Abhijit Gupta of ARAASTAT

    The first half of this course introduces general project and time management techniques. The second half focuses on best practices for the data science pipeline to minimize errors, maximize time to think, and maintain reproducibility.

Invited Sessions

  • Bayesian Adaptive Survey Designs, organized by Natalie Shlomo of the University of Manchester
  • Improper Imputation, organized by Paul T. von Hippel of The University of Texas
  • Recent Development of Bayesian Methods in Survey Sampling, organized by Yajuan Si of the University of Wisconsin-Madison
  • Session in Honor of Jim Lepkowski’s Retirement, organized by Michael R. Elliott of the University of Michigan
  • Analyzing Government Data with Missing Item Values: A WSS Invited Session, organized by Phillip Kott of RTI International
  • Environmental Surveys: A Hot Spot for Statisticians, organized by Stanislav Kolenikov of Abt SRBI
  • Multiple Imputation for Complex Health Survey Data, organized by Joseph Kang of the National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention
  • Nonparametric Saturated Methods to Handle Nonignorable Missing Data, organized by Mauricio Sadinle of Duke University
  • Recent Developments in Survey Sampling, Session in Honor of J.N.K. Rao’s 80th Birthday, organized by David Haziza of the University of Montréal
  • Using Big Data to Improve Official Economic Statistics, organized by Carma R. Hogue of the U.S. Census Bureau

Topic-Contributed Sessions

  • Current Themes in Record Linkage Research, organized by Jana L. Asher of AABB
  • Improving Efficiency and Maintaining High Data Quality: Outcomes for the 2017 Survey of Consumer Finances, organized by Catherine C. Haggerty of NORC at the University of Chicago
  • Methods for Imputing Missing Survey Data, organized by Daniell Toth of the Bureau of Labor Statistics
  • Multiple Imputation for Measurement Errors and Other Structured Patterns of Missing Data, organized by Philipp Gaffert of GfK
  • Practical Applications of Small Area Estimation, organized by Andreea L. Erciulescu of NISS and USDA NASS
  • New Developments in Small Area Estimation Research at the U.S. Census Bureau, organized by Robert Ashmead of the U.S. Census Bureau
  • Nonparametric Modeling of Survey Data, organized by Daniell Toth of the Bureau of Labor Statistics
  • Non-Probability Sampling and Estimation: Fit for Purpose Designs, organized by Karol Krotki of RTI International
  • Nontraditional Approaches for Sampling Rare Populations, organized by Sunghee Lee of the University of Michigan
  • Time-Trend Analysis with Complex Survey Data, organized by Dan Liao of RTI International
  • Advances in Modeling Multilevel Observational Data from Complex Surveys, organized by Mulugeta Gebregziabher of MUSC
  • GSS/SSS/SRMS Student Paper Award Presentations, organized by Stanislav Kolenikov of Abt SRBI
  • Improving Data Quality and Estimation Methods for the Current Employment Establishment Survey, organized by Greg Erkens of the Bureau of Labor Statistics

Contributed Sessions

  • Combining Data and Use of Administrative Lists
  • Estimation with Complex Samples
  • Estimation with Non-Probability Samples
  • Estimation with Statistical Models
  • Impact of Data Collection Modes and Data Sources
  • Imputation and Nonresponse Bias
  • Instrumentation and Data Quality
  • Predicting Attrition and Adaptive Strategies
  • Sample Design
  • Small Area Estimation and Use of Unit-Level Models
  • Weighting Adjustments
  • Weighting and Variance Estimation

P.M. Roundtables

  • The Connectivity of Data Science to Survey Design and Statistical Practice, led by Steve Cohen of RTI International

    This roundtable will focus on the capacity of data science to improve the design of surveys and their operations. We will also discuss strategies for reducing survey errors and enhancing data quality.

  • Election 2016 Polling: What We Learned, led by Mark Schulman of Abt SRBI

    We will cover both the methodological issues and the sizeable misses by many pollsters, as well as substantive issues in campaign strategies, electoral map, and media coverage.

The SRMS will hold its annual poster competitions in which the most informative and interesting posters will be awarded with cash prizes. The traditional poster session will be from 10:30 a.m. – 12:20 p.m. on August 1. SRMS has 11 contributed posters during this session.

The SRMS also gives out awards for winners in a speed presentation session that consists of 20 presenters. The speed presentation involves two parts. The first is for oral presentation and will take place from 8:30 a.m.–10:20 a.m. on August 2. Each presenter will talk for five minutes about their work. The second part is for “learning more.” It will immediately follow the oral session, taking place from 10:30 a.m.–12:20 p.m. Each presenter is provided a computer to present their work in detail.

If you are interested in volunteering as a judge for either the poster competition or speed presentation, contact SRMS program chair-elect, Stas Kolenikov. The winners will be announced at the section’s business meeting on the evening of August 2.

Jerzy Neyman (1894–1981)

Thu, 04/13/2017 - 7:00am
by Chin Long Chiang, Professor in the Graduate School, University of California, Berkeley


Jerzy Neyman, one of the principal architects of modern statistics, was Director of the Statistical Laboratory, University of California, Berkeley. He was born on April 16, 1894, into a Polish family in Bendery, Russia, and died on August 5, 1981, in Berkeley, California, at the age of 87. With Neyman’s passing, history closed a chapter on the early development of this important scientific field.

At the time of his birth, there was no Poland as a nation. “Poland proper” had been divided among Germany, Austria, and Russia. Neyman’s father was a lawyer. When Neyman was twelve years old, his father died of a heart attack. His caring mother moved her family to Kharkov, where he attended school and college. Although he was born a Pole, Neyman spoke Russian almost as early as he spoke Polish. At an early age, he could also speak Ukrainian, German, French, and Latin fluently. Upon his graduation from high school, through his mother’s arrangement, he joined a student group making a journey to see Europe outside Russia. Before entering the college in Kharkov, he decided to study mathematics instead of pursuing his father’s profession. He received his mother’s support and encouragement. “She had respect for intellectual activity,” Neyman fondly recalled to Constance Reid in the late 1970s. (Reid published her book entitled Neyman From Life in 1982.) In 1921, after a Polish-Soviet peace treaty, Neyman was sent to Poland in a repatriation of prisoners of war program between the two countries. Neyman saw his fatherland Poland for the first time when he was 27 years old.

Neyman’s interest in mathematics was reinforced when he studied with the Russian probabilist S. N. Bernstein at the University of Kharkov. When he read Henri Lebesgue’s Lecons sur L’intégration et la Recherche des Functions Primitives, Neyman was fascinated by sets, measure, and integration. During his college days he had proved five theorems on the Lebesgue integral on his own. His article entitled “Sur une théoréme metrique concernant les ensembles fermés,” published in 1923, was one of his early research papers in pure mathematics. His candidate thesis at the University of Kharkov (1916) was on the integral of Lebesque. In 1917, Neyman returned to the university for postgraduate study. In the following year he was a docent at the Institute of Technology, Kharkov. At the University of Warsaw, Neyman studied mathematics with Waclaw Sierpinski. He earned the Doctor of Philosophy degree from the University of Warsaw in 1924. The oral examination consisted of Rigorosum Major in mathematics and Rigorosum Minor in philosophy. No one knew more statistics than Neyman to examine him on the subject.

In the little spare time that he had during his student days, Neyman was heavily involved in teaching to earn a living. He also gave supplementary lectures for professors at the university and taught mathematics and statistics to college students.

Neyman first heard of Karl Pearson from reading Pearson’s book Grammar of Science (1892). Apparently, he was influenced by Pearson’s philosophical views as expressed in the book.

Neyman’s contact with statistics occurred early in his academic career. It appears that he had studied applications of mathematical statistics with Bernstein at the University of Kharkov. But he learned most statistics through work on his own, especially in agricultural experimentation. He held a position of “senior statistical assistant” at the National Agricultural Institute in Bydgoszcz, Poland, in 1921, and he was a special lecturer at the Central College of Agriculture in Warsaw in 1922.

In the fall of 1925, Sierpinski and Kazimierz Bassalik, the director of the National Agricultural Institute, were awarded a Polish Government Fellowship for Neyman to study mathematical statistics with Karl Pearson in London. Neyman was well prepared in mathematics and in statistics. While in London, Neyman and a young man about his own age, Pearson’s son Egon S. Pearson, became good friends.

During the academic year 1926–27, Neyman was on a Rockefeller fellowship to study pure mathematics in Paris. He attended lectures given by Emile Borel at the University of Paris and also lectures by Lebesgue and Jacques Hadamard at the College de France. In addition, he had some of his own notes read at these institutes. Quite possibly, the year of studying mathematics in Paris had prepared him well for his joint endeavor with Egon Pearson in the development of statistical theory in the years to come.

Neyman and Pearson’s joint work formally started in the spring of 1927, when Pearson visited Neyman in Paris. While there are no records of what transpired during the ten days during which they worked together, they must have laid out plans for their future joint project. At the end of the 1926–27 academic year, Neyman went back to Poland, and in 1928 he became head of the Biometric Laboratory at the Nencki Institute of Warsaw. He carried out his joint work with Pearson through correspondence between Warsaw and London. From 1928 to 1934, they published seven of their ten most important papers on the theory of testing statistical hypotheses.

In developing their theory, Neyman and Pearson recognized the need to include alternative hypotheses and they perceived the errors in testing hypotheses concerning unknown population values based on sample observations that are subject to variation. They called the error of rejecting a true hypothesis the first kind of error and the error of accepting a false hypothesis the second kind of error. They placed importance on the probability of rejecting a hypothesis when it is false. They called this probability the power of test. They proposed a term, ‘critical region’ to denote a set of sample statistical values leading to the rejection of the hypothesis being tested. The ‘size’ of a critical region is the probability of making the first kind of error, which they called the level of significance.

They called a hypothesis that completely specifies a probability distribution a simple hypothesis. A hypothesis that is not a simple hypothesis is a composite hypothesis. A hypothesis concerning the mean of a normal distribution with a known standard deviation, for example, is a simple hypothesis. The hypothesis is a composite hypothesis if the standard deviation is unknown.

It is now difficult for us to imagine how one could perform a statistical test without these concepts. But the Neyman-Pearson theory was a considerable departure from traditional hypothesis testing at the time. They were severely criticized for their new theory by the leading authorities of the field, especially by R. A. Fisher.

Neyman and Pearson used conceptual mathematics and logical reasoning to develop the theory of hypothesis testing. They emphasized “the importance of placing in a logical sequence the stages of reasoning in the solution of …inference.” In their initial papers (1928a) and (1928b), it seems that they were leading the reader, step by step, in their development of the theory. They relied on the concept of likelihood ratio in testing hypotheses concerning parameters in known probability distributions. And they elucidated their ideas further with specific examples and numerical computations.

After they had laid a solid mathematical foundation for their theory, they applied it to the problem of two samples (1930) and to the problem of k samples (1931). In one of their joint papers (1933) they used the likelihood ratio to establish an objective criterion for determining the best (in the sense of power of test) critical region for testing a simple hypothesis and a composite hypothesis. That was a high point of their accomplishment. The landscape of statistical hypothesis testing would no longer be the same.

In 1934, Neyman joined the faculty of E. S. Pearson’s Department of Applied Statistics at the University College London. From 1934 to 1938, they published only three more joint papers on testing hypotheses, possibly because of Pearson’s involvement in administrative responsibilities. Neyman, however, was still very productive during that period. From time to time, Neyman published papers on hypothesis testing on his own but most of the fundamental work was contained in his joint publications with Pearson.

When he was still in Poland, Neyman had developed the idea of confidence interval estimation. He even gave lectures on confidence interval estimation rather than hypothesis testing in his class at University College London in 1934. He published his work in 1937. At that time, many statisticians confused the confidence interval with the fiducial interval, a concept developed by Fisher. That confusion was soon dispelled by Fisher himself. Neyman clarified the difference between the two in his Lectures and Conferences (1938).

In addition to the theory of statistical inference, Neyman had made contributions to many other branches of statistics, such as the designs of agricultural experimentation (1923, 1925, 1935), the theory of sampling (1925, 1938, 1939), a class of ‘contagious’ distributions (1939), and others. He even used the “storks bring babies” example to show how to reach a wrong conclusion by misusing a correlation between variables, the so-called spurious correlation (1938).

Neyman’s work of applications of statistical methods in practical problems was very extensive. He considered practical problems as a source of inspiration for the theoretical statisticians.

There was an interesting feature in Neyman’s approach to practical problems. He had the ability to visualize the phenomena behind the data and a model of the mechanism that creates the phenomena. He would express the model in mathematical terms to produce new probability distributions, or new stochastic models. Only then would he find appropriate statistical methods to analyze the data on hand.

In the spring of 1937 Neyman delivered a series of lectures on mathematical statistics an probability at the Graduate School in the U.S. Department of Agriculture in Washington, DC. That was the first time that the American statistical public had the opportunity to hear statistical theory from Neyman in person. The lecture notes were subsequently published in 1937, and revised and expanded in 1952, under the title Lectures and Conferences on Mathematical Statistics and Probability. Among the reviews of the 1937 book, there was one written by William Feller, published in Zentralblatt, which reads in part as follows:

“The point of departure for the author is always actual practical problem and he never loses sight of the applications. At the same time his goal is always a truly rigorous mathematical theory. He appears to insist on absolute conceptual clarity and rigor, not only as a sound foundation, but also because it is really useful and necessary, particularly where the practical problem goes beyond the mathematical aspect…”

Feller’s words would apply equally well to Neyman’s other publications.

In 1938, Neyman accepted a mathematics professorship from the University of California at Berkeley. And he established the Statistical Laboratory, with himself as the director. That was the beginning of one of the preeminent statistical centers in the world. In 1955, Neyman established the Department of Statistics. He retained the title Director of the Statistical Laboratory.

Neyman was a very dynamic person, full of ideas and energy. Soon after the Statistical Laboratory was established and the teaching program was in good order, he began to plan a symposium of mathematical statistics and probability “to mark the end of the war and to stimulate the return to theoretical research.” The symposium had the participation of leading authorities in theoretical probability, in mathematical statistics, and in applied fields. The Proceedings of the symposium, edited by Neyman, were published in 1949 to “stimulate research and foster cooperation between the experimenter and the statistician.”

Success of the symposium prompted Neyman to plan a series of symposia, once every five years. The number of participants and the coverage grew from one symposium to the next. The Sixth Berkeley Symposium, held in three different periods in 1970 and 1971, was attended by 240 leading authors in 33 subject areas in theory of probability, in mathematical statistics, and in scientific fields with applications of statistics. The Proceedings, edited by LeCam, Neyman, and Scott, were published in 1972 in six volumes and 3397 pages—a gigantic undertaking.

These symposia supplemented the teaching programs and research academic activities normally carried out in universities and other academic institutions. They also had a great deal of influence on the attitude of theoretical statisticians and research scientists, making them recognize the need and the advantage of applications of statistics.

During the forty years that he was in Berkeley, Neyman had students come from all over the world to attend his lectures and to learn the proper way of conducting research. Neyman was a generous man. He helped students financially in any way he could. He recommended students for University scholarships and he secured federal grants for the support of students and faculty. At times, when he could not obtain the funds he needed to support students from any other sources, Neyman took the money out of his own pocket.

Neyman used to say “Statistics is the servant to all sciences.” In many ways Neyman had expanded the domain and improved the quality of the service.

Related Links

Jerzy Neyman”, School of Mathematics and Statistics, University of St. Andrews, Scotland

What Does Susan Hilsenbeck Do When She Is Not Being a Statistician?

Sat, 04/01/2017 - 7:00am
This column focuses on what statisticians do when they are not being statisticians. If you would like to share your pastime with readers, please email Megan Murphy, Amstat News managing editor.


Susan Hilsenbeck

Who are you, and what is your statistics position?

My name is Susan Hilsenbeck, and I am a professor of medicine and the leader of the Biostatistics and Informatics Shared Resource in the Dan L. Duncan Comprehensive Cancer Center at Baylor College of Medicine in Houston, Texas. I spend most of my time working collaboratively with investigators on cancer research.

Tell us about what you like to do for fun when you are not being a statistician.

When I am not working as a biostatistician, I like to scuba dive and I like to fish, but most of all I like to quilt.

Susan Hilsenbeck scuba dives.

What drew you to this hobby, and what keeps you interested?

I’ve been quilting on and off since 1982, which means I started shortly after the ‘quilt renaissance’ of the 1970s. I got started because I watched a program on PBS and I wanted to make a present for a friend having a baby. Even for my first quilt, I aimed for something original, although I am a strong believer in mastering good technique and then branching out to make it your own.

I work pretty slowly and savor every aspect of the process. I’ve made very few bed-sized quilts and focus instead on baby-sized and wall art quilts.

One of the things I love about quilts is how the gift of a quilt expresses love and regard. The experience of making a group quilt is especially fun and can help build bonds of friendship. Over the last 17 years, since joining Baylor, my co-workers and I have made more than 30 lap/crib-sized group quilts to celebrate life events like babies, marriages, retirements, etc. The group quilts are usually more whimsical in style, while my solo quilts are more on the experimental and art side (or at least I like to think so).

Susan Hilsenbeck included a DNA molecule applique in this quilt she designed.

The thing I like best about quilting—as a mode of expression—is the blend of analysis/engineering to figure how to put something together; the physical skill needed to execute the design; and the freer, artistic side of design, composition, color, etc. It seems like the perfect mix for all sides of my brain. There is also something incredibly relaxing and centering about spending a couple of hours focused on stitching. I keep at it because there is always more to learn and the possibilities are endless.

To view some of Hillsenbeck’s work, visit her blog.

2017 Data Challenge Sees 16 Contestants

Sat, 04/01/2017 - 7:00am

The ASA’s Statistical Computing, Government Statistics (GSS), and Statistical Graphics sections are sponsoring the 2017 Data Challenge, which will take place at the Joint Statistical Meetings in Baltimore.

The goal of the contest, which began earlier this year with analyses of the Bureau of Labor Statistic’s Consumer Expenditure Survey, is to challenge participants to analyze a government data set using statistical and visualization tools and methods. Of the college students and professionals who submitted an analysis, 16 were chosen to present their results in a speed poster session at JSM. All JSM attendees are encouraged to view the presentations.

There will be two award categories: professional (one level) and student (three levels). These awards will be announced at the GSS general membership meeting.

The sections are also moving forward with a special issue of Computational Statistics, which will feature refereed articles from contestants in the 2016 Data Challenge. The data set for the 2016 challenge came from the Department of Transportation’s General Estimates Systems. For the 2017 challenge, the plan is to publish selected refereed articles in a special issue of the Monthly Labor Review.

Master’s Programs in Data Science and Analytics

Sat, 04/01/2017 - 7:00am
More universities are starting master’s programs in data science and analytics due to the wide interest from students and employers. Amstat News reached out to the statistical community involved in such programs. Given their interdisciplinary nature, we identified those that involved faculty with expertise in different disciplines to jointly reply to our questions. In 2015, for example, the ASA issued a statement about the role of statistics in data science, saying statistics is one of three foundational disciplines of data science. While the ASA has not issued a statement about the role of statistics in analytics, we assume statistics to also be foundational there. For this reason, we highlight the programs that are cross-disciplinary and engage statisticians. We will publish responses over a few issues of Amstat News. ~ Steve Pierson, ASA Director of Science Policy


University of Tennessee

Robert Mee is the William and Sara Clark Professor of Business, Department of Business Analytics and Statistics, Haslam College of Business, University of Tennessee. He is an ASA fellow who earned his PhD in statistics from Iowa State University.


Master’s in Business Analytics

Year in which first students graduated: 2011
Number of students currently enrolled: 38 full time, 6 part time. In the fall semester, approximately 80 (first- and second-year combined)

How do you view the relationship between statistics and data science?
Statistics informs both data collection and analysis. Other disciplines are involved with acquiring, managing, and analyzing data, but statistics gives particular attention to potential biases in both collection of data and in estimates of models. Statistics has the tools for quantifying uncertainty, understanding sources of variation, and confirming or contradicting hypotheses. Data science is centered on data and algorithms, as opposed to statistics, which begins with a problem to be addressed.

Describe the basic elements of your data science curriculum and how it was developed.
The core MSBA curriculum combines statistics, data mining/machine learning, optimization, database, and other computing skills, as well as giving the students a foundational understanding of the problems businesses address with analytics. Our students can choose electives in statistics, customer analytics, supply chain analytics, machine learning, or computer science to prepare them for their intended career direction.

We include a business perspective in our curriculum by maintaining close relationships with members of our Business Analytics Forum. In addition, many MSBA faculty members have extensive consulting and executive MBA teaching experience.

What was your primary motivation(s) for developing a master’s data science program? What’s been the reaction from students so far?
In 2009, our visionary department head brought an IBM white paper on the future of analytics to the attention of faculty. The paper emphasized applied statistics, business intelligence, and process optimization. At that time, we had two parallel master’s programs, one in statistics and the other in management science. We decided to combine these programs, creating a business analytics MS that now includes applied statistics, data mining, optimization, and Big Data tools.

What types of jobs are you preparing your graduates for?
Data scientist, analytics consultant, data analyst for supply chain or marketing. The title business analyst is not quite suitable, since this is often the title intended for less-quantitative MBA graduates.

What advice do you have for students considering a data science degree?
Take engineering calculus and learn some programming language. Pursue business analytics if you want to solve quantitatively-oriented problems and enjoy working with vast amounts of data to produce actionable insights that affect a business’s bottom line.

The letter “T” is sometimes used to characterize the business analytics masters, with the top of the T reflecting the breadth of this interdisciplinary degree and the vertical part indicating depth in one technical area. A statistics or computer science MS typically would have greater depth, but would lack the breadth of an MSBA.

Describe the employer demand for your graduates/students.
Last December, we graduated our 6th class of MSBA students. Through 2015, we have had 100% placement within three months of graduation. Companies making three or more hires include Amazon, The Boeing Company, Eastman Chemical Company, Hanesbrands Inc., Home Depot, and Regal Entertainment Group. We have also had many graduates work for consulting companies, including Accenture, Deloitte, EY, McKinsey & Company, KPMG, PWC, and several smaller firms.

Do you have any advice for institutions considering the establishment of such a degree?
Know your competitors. Consider the target incoming students, as this determines the length of the program. Consider your strategic advantages, especially ties with industry.

George Mason University

Robert Osgood is the director of the data analytics engineering program. He has expertise in developing and applying analytics to law enforcement in digital forensics, enterprise case management, cyber crime, counterintelligence, information security/technology, team leadership, and critical infrastructure protection.


Daniel Carr is professor of statistics and director of the statistics concentration in the data analytics engineering program. His driving interest is to create statistical graphic designs and software to address constraints posed by human cognition and challenges posed by new kinds of data and large data sets.


Master’s in Data Analytics Engineering

Year in which first students expected to graduate: 2014
Number of students currently enrolled: 290
Partnering departments: Volgenau School of Engineering (Lead): Statistics, Systems Engineering and Operations Research, Information Sciences and Technology, Computer Science
Program format: The MS Data Analytics Engineering (DAEN) program is an in-person program, although some courses are offered online. Our student body is a mix of full-time and part-time, both domestic and international. We are actively involved with our corporate partners and career services unit to offer internships for students.

Please describe the basic elements of your data science/analytics curriculum and how the curriculum was developed.
The MS DAEN program is interdisciplinary. It revolves around a 15-credit core component with a 15-credit concentration component for a total of 30 credits.

Concentrations include applied analytics, business analytics, data mining, digital forensics, health care analytics, predictive analytics, and statistics. The core component consists of four courses, each taught in different departments, and a capstone.

Our admission criteria vary by concentration, but a minimum of one semester each of calculus, programming, and statistics is required. For the statistics concentration, three semesters of calculus, linear algebra, and probability are required.

What was your primary motivation(s) for developing a master’s data science/analytics program? What’s been the reaction from students so far?
The Volgenau School of Engineering at Mason, through working with its corporate partners and its advisory board, identified the need for data analytics education. Our investigation showed there was a growing demand for individuals with data analytics skills. Northern Virginia is a particularly appropriate area for employment in data analytics.

Student reaction has been extremely positive. Our spring 2017 enrollment of 290 students shows a definite demand for data analytics knowledge.

How do you view the relationship between statistics and data science/analytics?
There are two main components to data analytics: computing technology and statistics. Statistical analysis is what drives meaning from the massive data sets deposited with computing technology (hardware and software). The statistical component also includes visualization and data reduction.

What types of jobs are you preparing your graduates for?
Students are obtaining positions in a varied array of industries: cyber security, finance, government, information and knowledge management, software development, and e-commerce. Wherever data are collected, the need for analysis and statistical techniques is present.

What advice do you have for students considering a data science/analytics degree?
All departments (disciplines) look at the world from a certain point of view. Data analytics is not just computer science, or statistics, or business processes. It’s all of the above. So students looking for a more interdisciplinary view must seriously consider data analytics versus one of the traditional degrees.

It needs to be pointed out that data analytics requires more than just novice knowledge of com-puter science and statistics. For example, students need a solid statistical foundation normally found in a typical undergraduate statistics class. Also, knowledge of probability is also quite helpful. A minimum of one semester of calculus is also critical. Students need programming expertise. Any language will work, but Python is a particularly valuable language. Knowledge of database design and interaction is also desirable. Development of communication skills is essential in dealing with interdisciplinary stakeholders.

With these core components in place, a student can leverage his/her data analytics learning experience at Mason to the fullest extent.

Describe the employer demand for your graduates/students.
Our interaction with our corporate partners and exit survey data show a significant demand for Mason graduates. The finance, government, and technical sectors have employed our data analytics graduates.

Do you have any advice for institutions considering the establishment of such a degree?
The program must be interdisciplinary. No one owns data analytics, but everyone uses data analytics, so everyone needs to be a stakeholder. At Mason, the department of statistics is housed in the engineering school, along with the department of computer science, department of information sciences and technology (IT management), and department of systems engineering and operations research (predictive analytics). While each department created its own concentration with its own set of prerequisites, a common set of core courses was developed with minimal prerequisites, and each department contributed a course to that core.

University of Minnesota

Cavan Reilly earned his PhD in statistics at Columbia University in 2000, after which he joined the faculty in biostatistics at the University of Minnesota, where he has remained. Over the last several years, he has transitioned to working on more applied problems with an emphasis on clinical research on infectious diseases.


Dan Boley is professor of computer science and director of the graduate studies for master’s of science in data science program at the University of Minnesota. His research interests include computational methods in linear algebra, scalable data mining algorithms, algebraic models in systems and evolutionary biology, and biochemical metabolic networks.


Master’s in Data Science

Year in which first students expected to graduate: 2017
Number of students currently enrolled: 34
Partnering departments: Computer Science and Engineering (Lead), Statistics, Public Health (Division of Biostatistics), Electrical and Computer Engineering
Program format: Combination/ (distance learning option available)/typically traditional full-time, but a part-time option is available. Assistantships are available, but not guaranteed. Thirty-one credit hours required; six credit hours over two semesters are for a cumulative research project.

Please describe the basic elements of your data science/analytics curriculum and how the curriculum was developed.
Our data science curriculum is intended the fill the spaces between algorithmics, statistical analysis, and modern computing infrastructure (these are the three core areas of our program). Finding the appropriate balance between these components and not simply duplicating and re-labeling existing opportunities for students was our guiding principle.

The curriculum was developed jointly by faculty from computer science and engineering, the school of statistics, electrical engineering and computer science, and the division of biostatistics in the school of public health. Through a series of meetings open to anyone interested, a consensus emerged that our program would focus on providing students with rigorous training in statistical methodology combined with a practical focus on computational feasibility in the age of Big Data, informed by a contemporary understanding of the possibilities engendered by the latest developments in hardware.

To distinguish our degree from the more business-oriented degree more commonly offered by many institutions around the country (including ours), we opted for a rigorous degree demanding a solid background in computing (3–4 semesters) and math/stat (3–4 semesters) at the undergraduate level as a minimum requirement. The other essential ingredient in our degree is a research component consisting of a two-semester capstone project.

What was your primary motivation(s) for developing a master’s data science/analytics program? What’s been the reaction from students so far?
This degree was created to meet the demand from companies for practitioners with solid training in scalable computing methods combined with a solid understanding of statistical issues and methods. Individual programs already had solid curricula in place to address various components of data science, but it was hard for students to package a program with the necessary elements without a cross-disciplinary degree. Many students in all contributing programs were already trying to assemble plans of study that would provide them with the necessary expertise one would expect from a degree in data science; this program has formalized that training. The response from students thus far has been very positive.

How do you view the relationship between statistics and data science/analytics?
We see statistics as but one component of a well-balanced program in data science. Some classical statistical techniques that served science well in the 20th century are simply not up to the task of dealing with data sets from the 21st century. This has led to extensive cross-fertilization between topics traditionally viewed as more within the realm of computer science and electrical engineering than statistics and ideas of great interest to statisticians (e.g., machine learning and causal inference). While such developments are positively affecting the practice of statistics and computer science, there are still opportunities to advance science that require perspectives and skills beyond what is possible in the context of a statistics curriculum or a computer science curriculum. As such, we see the relationship as complementary.

What types of jobs are you preparing your graduates for?
This is a new program, but our students have found internships at various companies employing sophisticated technology in the area, including health insurance, retail, and major social media companies on the coasts.

What advice do you have for students considering a data science/analytics degree?
Such students should strive for a balance between the three core areas identified above during their undergraduate education. Less balance is appropriate for a student specializing in computer science or statistics. The usual calculus sequence and linear algebra are still essential. A year of probability and statistics and a year of data structures and algorithms are becoming prerequisites. An internship is always helpful.

If a student is interested in graduate-level training in a field involving machine learning, data analytics, artificial intelligence, or data mining, having a solid computing background will be essential to implement anything novel. However, a solid statistics background will be essential to ensure that whatever is implemented gives statistically reliable predictions. Many employers have realized that a computing background alone is not sufficient for their next level of data systems development.

Do you have any advice for institutions considering the establishment of such a degree?
Ensure there is no obvious path for students to accomplish the same Stat/CS curriculum through an existing degree (unless you are prepared to simply re-label the existing degree). Make a choice between a regular graduate program with a choice of courses from a short list of requirements on the one hand and a cohort program where all students take the same courses together in sync.

Bentley University

Mingfei Li is an associate professor at Bentley University, where she has been a faculty member since 2008. She is currently serving as the MSBA program director and coordinator of the business analytics certificate and concentration programs. Her research interests include health analytics and sequential predictions.


Master of Science in Business Analytics

Year in which first students graduated: 2016
Number of students currently enrolled: 142 for the MSBA degree, 95 for a business analytics certificate or concentration within other degrees
Partnering departments: Mathematical sciences
Program format: Bentley University is a private business university broken down into a school of business and a school of arts and sciences. The master of science in business analytics (MSBA) is the only graduate degree offered by a department (mathematical sciences) not in the school of business. The department of mathematical sciences provides the quantitative curriculum for the MSBA degree, as well as all the other analytic courses offered at Bentley in statistics, mathematics, quantitative finance, data mining, Big Data, and operations research. Consequently, the department of mathematical sciences is, by necessity, interdisciplinary.

Currently, our program is an in-person program, but has synchronous hybrid classes for some of the courses. The MSBA degree requires 30 credits, and many courses require student projects. We have both full-time and part-time students. Bentley also provides scholarship and assistantship for some outstanding students.

Please describe the basic elements of your data science/analytics curriculum and how the curriculum was developed.
Our program has introductory statistics as a program prerequisite. Students are not required to have coding/programming skills upon entry because we offer programming classes from our curriculum: SQL, data science for R programming, reporting and data visualization, Java programming, HTML, Hadoop, MapReduce introduction, etc.

The MSBA has six required classes: SQL, operations research, and four business-oriented applied statistics classes. Students can choose four elective classes from a list, which includes classes from disciplines representing areas of business applications such as computer science, marketing, finance, economics, management, and informational process management. We collaborate with other departments on these elective classes.

For topics like database development and management, we use courses offered by our computer science colleagues. For business context, we use courses offered by departments of the school of business such as finance, marketing, management, economics, and information management. So while the MSBA program is highly interdisciplinary across both schools in our university, the core focus on analytics led to the decision that the MSBA be managed in the mathematical sciences department.

Beyond the existing curriculum, we continue to develop additional analytical courses to enrich the program curriculum such as machine learning (using R and Python) and design of experiments for business.

What was your primary motivation(s) for developing a master’s data science/analytics program? What’s been the reaction from students so far?
Bentley began to offer analytic courses in the 1990s. In 2006, we began to offer a certificate program in business analytics in response to a strong demand for graduates with comprehensive skills centered on statistics and including computer sciences skills, business knowledge, and communications skills. With strong demand from both students and employers, we launched the MSBA degree program in 2013.

Our enrollment rapidly increased from our first class of 40 to the current 162 students. Applications have increased in a similar manner.

How do you view the relationship between statistics and data science/analytics?
Applied statistics is the core of both data science and analytics. Understanding data and knowing how to analyze data are essential in most of applications. Computer science knowledge and programming skills are necessary to facilitate most analyses. Understanding the business questions and stakeholders’ interests provide the guide to statistical thinking and planning for analysis. Therefore, in our program, we require that students develop competency and skills in all three areas: analytics (statistics and operations research), computer science (related to data management and computing), and business (context for application).

What types of jobs are you preparing your graduates for?
Our graduates are hired by companies across different sectors, both private and nonprofit. Because of the wide applicability of business analytics, graduates have a variety of job titles such as business system analyst, data scientist, analytics consultant, data analytics adviser, research analyst, senior modeling analyst, business analyst officer, and business analyst. Employers of our students include CVS, National Grid, Toys R Us, Accenture, Deloitte, Ernst and Young, EMC, and Boston Children’s Hospital.

What advice do you have for students considering a data science/analytics degree?
Compared to a computer science degree or a traditional statistics degree, Bentley’s MSBA is a degree that integrates applied statistics with computer science, operations research, and business knowledge. Through this degree, students get interdisciplinary analytic knowledge with a practical understanding of business. Graduates can work with both business and technical teams to be a problem solver and innovator, providing decision support and business insights.

Describe the employer demand for your graduates/students.
There is a strong demand from employers for our MSBA students. From a survey of our first MSBA graduating class (88% response rate), 94% of graduates got full-time jobs within 90 days of graduation. The average annual salary for these graduates was $78,000, and the median salary was $80,000.

Do you have any advice for institutions considering the establishment of such a degree?
Being an interdisciplinary program, the MSBA needs support from multiple academic and administrative departments. Communication is crucial for the coordination and management of the program, as well as for curriculum development. A designated program director has been critical to coordinate all aspects of the program—from admissions to advising to placement—and to work directly with faculty colleagues, administrators, and students.

Students enrolling in the MSBA program have varied backgrounds, interests, and future goals. It is challenging to accommodate this degree of student variation in the curriculum, academic advising, and career services. The program director interacts with students personally throughout their enrollment in the program to understand their individual background and interests and advise their course selections and career preparation to help students achieve their personal goals.

Meet Hubert Hamer: NASS Administrator

Sat, 04/01/2017 - 7:00am
Amstat News invited Hubert Hamer—administrator of the National Agricultural Statistics Service—to respond to the following questions so readers could learn more about him and the agency he leads.

A graduate of Tennessee State University, Hubert Hamer once served as director of the Statistics Division of the USDA, which produces and releases more than 400 national statistical reports each year covering the agency’s crops, livestock, economic, demographic, environmental, and census programs. He also served as executive director of the NASS Agricultural Statistics Board and executive director of the Advisory Committee on Agriculture Statistics.


What about this position appealed to you?

It has always been a goal of mine to provide leadership for the organization I grew up in. I’ve always loved and understood the importance of our mission to provide timely, accurate, and useful statistics in service to agriculture. I think it is rare that one has the opportunity to build a career in something meaningful that one really cares about. I feel very fortunate that I have been able to do just that by combining agriculture and statistics. By working my way up through the ranks at NASS in the Washington, DC, headquarters and in offices around the country, I’ve learned from my colleagues, supervisors, and mentors in NASS and USDA. I’m honored to be able to apply that knowledge and experience to lead this agency.

Describe the top 2–3 priorities you have for the National Agricultural Statistics Service.

Service and our commitment to U.S. agriculture are what drive us at NASS. To that end, I have three areas of focus:

  • Relationships with survey respondents
  • NASS employees
  • Advancing our use of technology to ensure data quality and usability

In my first months as administrator, I have made it a priority to get out and meet with staff, stakeholders, and respondents across the country. My purpose for these meetings is to strengthen the dialog and relationships to remain relevant and provide the outstanding products and services the agriculture community and others expect from us.

In turning to our staff, my goal is to have an environment in which our outstanding employees can do their best work. This includes a safe and inclusive workplace with diversity of people and ideas; where vigorous, respectful debate is encouraged; and where employees can continue to harness their talent and work ethic to fulfill both their career goals and our agency mission. My expectation is that everyone who interacts with NASS staff finds us to be helpful, pleasant, and professional.

In the area of technology, we have amazing tools to help us that I couldn’t imagine using even just a few years ago. I think George Washington, one of the earliest compilers of agricultural data, and those who later produced the first census of agriculture in 1840 would be astounded. Like other statistical organizations, we are working hard to efficiently and effectively harness geospatial tools, a wide range of available data, the internet, the cloud, data visualization, and many other constantly evolving assets to enhance data collection, analysis, and dissemination. To me, technology is our future.

What do you see as your biggest challenge(s) for NASS?

Probably the three greatest challenges for us today are reflected in my priorities. I believe our greatest challenges are keeping up with expanding data needs, reversing the trend of declining response rates that many survey-based organizations are experiencing, and bringing along a next generation of statisticians.

Starting with the topic of meeting data needs … with interests as varied as research and education, community-based planning, farm-related marketing, commodity markets, and the Farm Bill, we are constantly looking forward. For example, in the past few years, we’ve expanded our portfolio of publicly available data to include topics as diverse as grain crushing, flour milling, local foods, organic production, and the cost of pollination. We’ve also created a host of new types of data products. On obtaining sufficient response to surveys, I see this as a real challenge and an opportunity. We are reinvigorating our relationships with farm organizations to see if they can help us reach out to the farmers and ranchers from whom we request information. This has been a very positive effort on many levels.

Finally, one of the great things about NASS is our sense of family. Like me, many come out of college and stay here for their whole career. We have a large number of people who have been in the agency for decades and are starting to retire. We have a renewed focus on recruiting young people, training them, and keeping them so we maintain our knowledge and top-notch skill base.

What kind of support from the statistical community do you look for?

The statistical community has always been an incredible resource for sharing experiences, technologies, and best practices, as well as for looking ahead for new developments. We are also a great support for each other, which I always find valuable. A couple of areas I expect we’ll be focusing on together are to understand and address the reasons for declining response rates, educate decision makers and stakeholders across disciplines about the value of data as a public resource, and collaborate on and use new technologies that enhance our data-collection and dissemination practices, so we all remain relevant.

Prior to your tenure, what do you see as the biggest recent accomplishment of the agency?

This may sound routine, but maintaining our schedule of releasing some 450 reports a year on time and without errors. It is truly a testament to our staff’s commitment, especially while keeping up with new technologies and learning to collect data on new topics from farmers, ranchers, and agricultural businesses they may never have worked with before.

And along those lines, of course, it is a major achievement to conduct a successful Census of Agriculture every five years and create a portfolio of new customer-centric data products. We’ve been working since the release of the 2012 Census of Agriculture to get ready for the 2017 data collection, which will begin late November with a mail out to 3.1 million farmers and ranchers. We are really excited about a new online survey data-collection tool that we expect to use the first time for the census. We are testing it now and have great hopes that respondents will find it to be a convenient and flexible way to fill out the Census of Agriculture.

CSP 2017 Brings Statisticians Face to Face

Sat, 04/01/2017 - 7:00am

Tim Hesterberg gives a short course.

Sara Burns (winner of the Bartko scholarship) with John Bartko

Bei-Hung Chang presents her poster at the Opening Mixer.

Curtin Award winner Jami Jackson Mulgrave presents her poster at the Opening Mixer.

Roland Albert Matsouaka, of Duke University, and Peter Hanging Zhang, of Otsuka Pharmaceutical Development and Commercialization, Inc.

David Banks gives the keynote address.

The audience listens to David Banks during the keynote presentation.

Hadley Wickham answers a question from Sara Burns during his short course, “Expressing Yourself with R.”

Photos courtesy of Meg Ruyle/ASA

Moon Jung Cho, Bureau of Labor Statistics and CSP Program Chair


The 2017 Conference on Statistical Practice was held February 23–25. Four hundred fifty participants gathered at the Hyatt Regency Jacksonville Riverfront in Jacksonville, Florida. 

The conference began with short courses and an opening mixer on Thursday. The following two days were filled with presentations, poster sessions, tutorials, and practical computing demonstrations. Keynote speaker David Banks set the tone of the conference with his presentation “Snakes and Ladders: Challenges in Forging a Career in Statistics.”

Unique to CSP is its theme, “Communication, Collaboration, and Career Development.” This theme provides participants with tools for leadership and external communication, with the goal of empowering participants to bring a positive impact to their organizations.

A new feature this year was a face-to-face task group charged with facilitating networking for conference attendees that allowed them to engage and connect with each other easily. Some of the face-to-face ideas included the following:

    SPEED NETWORKING: We set up speed networking tables at the poster session Friday evening. Participants were assigned a table for a quick round of speed networking to meet and discuss the conference. 

    COMMON AREA GATHERING SIGNS: Gathering signs were placed in the common area to encourage participants to have meeting points prior to going for coffee, lunch, dinner, and other conference-related events.

    DINNER FOR CONSULTANTS: There was a dinner for aspiring and practicing statistical consultants. The goal was to make personal and professional connections and get support from colleagues.

    THEME DINNERS: Participants signed up for topic-based group dinners on Thursday and Friday evenings. The “Data for Good” theme dinner was led by David Corliss; “Misleading Graphs” by Naomi Robbins; and “Organizational Impact, Analytics Change, and Soft Skills for Success” by Terri Henderson.

    EVENING OUTING: Around 30 people participated in a Jacksonville “Legends and Liars” walking tour led by a local historian and storyteller Friday evening.

As in previous years, we continued the CSP Mentoring Program, which was designed to establish a 1:1 mentoring relationship between junior and senior statistical practitioners and provide an opportunity to enhance personal and professional development goals.

By design, the CSP space is centered on a common area that holds the exhibitor booths, opening mixer, poster sessions, and breakfast and any refreshments. This physical arrangement, along with the small conference size, provides a great opportunity for continuous networking throughout the conference.

The CSP 2017 Best Student Poster Award went to Thomas Metzger of Virginia Tech for “Detecting Interaction in Two-Way Unreplicated Experiments via Bayesian Model Selection” and Carl Ganz of UCLA Center for Health Policy Research for “Using Shiny to Efficiently Process Survey Data.”

CSP 2018 will take place in Portland, Oregon. We hope to see you there!

Why Be an Independent Consultant?

Sat, 04/01/2017 - 7:00am
This column is written for anyone engaged in or interested in statistical consulting. It includes articles ranging from what starting a consulting business would entail to what could be taught in a consulting course. If you have ideas for articles, contact the ASA’s Section on Statistical Consulting publication’s officer, Mary Kwasny.

Stephen Simon is a part-time independent statistical consultant and part-time faculty member in the department of biomedical and health informatics at the University of Missouri-Kansas City. He writes about statistics, evidence-based medicine, and research ethics.


So, you want to be an independent statistical consultant? Hang out your shingle and start helping people who come your way? Give up all the security that comes with consulting within a larger organization?

Are you crazy?

Inside a large organization, you have a support network. You have a human resources department that can help you update your insurance coverage when you get married. You probably have access to an administrative assistant who can help you prepare expense reports. That large organization will support your professional development, paying your way to the continuing education course at the Joint Statistical Meetings.

Independent consulting is not a job for the timid. But if you like being in control, it’s the best job in the world.

I like to write books and articles, and I haven’t figured out yet how to bill any of this work to a particular client. That’s true in some large organizations as well, but many places do offer time and support for professional activities that are not directly tied to a particular client.

In a large organization, if you don’t know how to run a mediation analysis, you can walk down the hall to a colleague’s office to ask a few questions. You also don’t have to go out and find customers, because your customers are down the hallway from you, as well.

At a larger organization, you will have a boss and, for all we like to gripe about bosses, they can often be a great benefit to your career. They review your work and make suggestions on how to do better. They pick the work assignments they think will help you grow and become more valuable to your organization. They counsel you when you have problems.

Most importantly, if you’re part of a larger organization, your paycheck and working hours stay constant during busy and quiet times. In contrast, it’s feast or famine in the world of independent consulting.

As an independent consultant, you end up doing a lot that isn’t really statistical in nature and may take you outside your comfort zone. I work with an accountant, but I’ve had to learn a lot more about accounting than I expected to. I’m the sort of person who, if the amount in your checkbook register is within a hundred dollars of what the bank says you have, thinks the difference must be sampling error.

It’s not just accounting. I pay someone to do my taxes, but when I became an independent consultant, the amount of paperwork I had to pull together for him by April 15 tripled.

You can get legal advice on the best type of business entity to set up, but this is your business, so you will still need to understand the fundamental differences between a sole proprietorship, limited liability corporation, S corporation, and C corporation. Even if your lawyer reviews your contracts, you will end up reading them in detail before you sign off.

Most importantly, you don’t have an organization that finds work for you. So, if you don’t market yourself properly, you won’t have any clients, you won’t make any money, and you will starve to death.

And yet, if you’re the right person, you’d be crazy not to consider a career as an independent consultant. When you are your own business, you have a level of control that is liberating. You don’t like a particular client? You have the option of just walking away. It’s a loss of income, for sure, but some clients are not worth any amount of income. If you try to walk away from a client at a larger organization, your boss needs to okay it first. Nine times out of 10, you will get so much grief about not being a “team player” that it won’t be worth it.

Independent consulting is indeed spread irregularly, but even during busy times, you still have a lot of control over when and where you do your work. That was one of the biggest attractions for me. I have the option of going on field trips with my son and attending all of his track meets. I can be home with him when he’s sick and take him to all his doctor’s appointments. And when he sleeps until 2 p.m. on weekends, that’s when I get much of my work done.

As an independent consultant, you do have to pay for your own continuing education, but the nice thing (beyond it being a tax deduction) is that you don’t have to justify to anyone other than yourself that it’s about time you learned how to run all these new Bayesian models.

That support network in a larger organization? It’s not always there, to be quite honest, and you can build your own support network as an independent consultant. I volunteered to step in as president of the Kansas City R Users Group, and beyond the exposure and the number of new clients it has brought me, the other members of this group have been invaluable resources for things like version control software, data mining, and text analytics.

Independent consulting is not a job for the timid. But if you like being in control, it’s the best job in the world.

BEA’s Innovation Spurs Projects for Richer Economic Statistics

Sat, 04/01/2017 - 7:00am

Brian Moyer oversees the Bureau of Economic Analysis’ production of official economic statistics, which provide a comprehensive, up-to-date picture of the U.S. economy that aids in decision making by businesses, policymakers, and households. He holds a PhD in economics from American University.

To continue capturing a full and detailed picture of a dynamic, $18 trillion-plus economy, the people of the Bureau of Economic Analysis (BEA) have to be economic data pioneers. They are committed to innovating and exploring, whether the bureau is enhancing existing statistics or creating new ways to measure the U.S. economy. That mind-set is crucial to delivering on BEA’s mission: producing the timeliest, relevant, and accurate economic statistics for the American public in an objective and cost-effective manner.

Here is a snapshot of a few of the data projects BEA’s economists are working on.

The Digital Economy. We are moving forward on a three-pronged plan to better measure fast-changing technologies and their effect on the U.S. economy. One focus is refining price measurements to better capture innovations in high-tech goods and services such as software, cellphones, personal computers, computer servers, cloud computing, and medical imaging equipment.

To tackle improvements in quality-adjusted price measures for such products, BEA is doing the following:

  • Conducting an in-house review of GDP and its components to identify areas in which existing quality-adjusted prices could be improved or new indexes could be introduced
  • Partnering with source data agencies, including the Federal Reserve Board and Bureau of Labor Statistics, to improve software and medical equipment price measurement
  • Engaging experts for specialized research such as building new price indexes for cloud computing

In addition, BEA economists are developing a roadmap to define and measure the digital economy. BEA is researching how to more accurately measure the impact of information technology on the overall U.S. economy and how to improve the measurement of digitally enabled commerce.

Third, BEA is researching the economic impact of “free” entertainment such as Facebook apps and internet games, which are largely supported by advertising revenue. And BEA is committed to better understanding the impact of technology-enabled, peer-to-peer access to goods and services—typically referred to as the “sharing economy.”

Health Care. Created in 2015, our new set of health care statistics break out spending by the treatment of disease, such as circulatory diseases or cancers, rather than by the place of service, such as a hospital or doctor’s office. Each year, BEA plans to release a fresh batch of health care statistics, building a longer time series. Data are currently available for 2000 through 2013. Figures for 2014 will be released later this year.

These data offer new insights into health care, which accounts for about 18 percent of the U.S. economy. After years of research, BEA created a “blended account,” which combines data from multiple public and private sources, including large claims databases covering millions of enrollees and billions of claims.

In its next steps, BEA plans to research linking changes in the costs of treating diseases to improvements in the quality of treatments, including advances that lead to better health outcomes. That’s one of the biggest challenges in precisely measuring medical spending and prices—not unlike what BEA is confronting in the high-tech sector.

BEA also plans to build a detailed input-output framework for health care spending, giving users a way to better analyze the production of goods and services by health care industries. BEA will incorporate prices that reflect the costs of treating diseases into the input-output framework.

There’s More. We have other innovative data projects in the wings, including laying the groundwork for a new set of statistics—a small business satellite account. It would measure, for the first time, the size and health of a sector that’s often at the leading edge of risk-taking, entrepreneurship, and economic growth in the United States.

BEA is also exploring the feasibility of measuring economic growth in the nation’s 3,000-plus counties. These first-of-their-kind BEA statistics would help businesses identify local markets for their products, assist local governments seeking to attract investment, and give a fuller picture of the U.S. economic landscape.

On the global front, BEA is working to expand statistics to provide a more detailed look at how businesses buy and sell services around the world. Quarterly statistics on U.S. trade in services will be expanded to cover 90 countries and country groups (from the current 38.) Details will be published about some of the most dynamic sectors, including research and development, intellectual property, and medical services.

Data Tools. We are also creating new ways to access our data. The newest offering makes data available through the bea.R Library, an open-source data tool for users of the statistical programming language “R.” This gives users a quick way to access our economic statistics, requiring only a few lines of code to do so. BEA’s data also is available through its application programming interface (API), interactive data tables, and other data tools at the BEA website.

Biometrics Section Prepares for JSM 2017

Sat, 04/01/2017 - 7:00am
Edited by Zheyu Wang, Biometrics Section Publications Officer

Want to get more involved in the Biometrics Section? Interested in contributing articles to the Biometrics Section newsletter? Contact the section’s publication officer, Zheyu Wang.

JSM 2017 Program

The Biometrics Section will sponsor the following continuing education (CE) courses at the 2017 Joint Statistical Meetings in Baltimore:

    Longitudinal and Incomplete Data
    Instructor(s): Geert Molenberghs and Geert Verbeke

    An Introduction to the Joint Modeling of Longitudinal and Survival Data, with Applications in R
    Instructor(s): Dimitris Rizopoulos  

    Bayesian Designs for Phase I-II Clinical Trials
    Instructor(s): Peter Thall and Ying Yuan

    Regression Modeling Strategies
    Instructor(s): Frank Harrell

    Precision Medicine Through Optimal Treatment Regimes
    Instructor(s): Eric Laber, Marie Davidian, Anastasios (Butch) Tsiatis, and Shannon Holloway

    Statistical Analysis with Missing Data
    Instructor(s): Roderick Little and Trivellore Raghunathan

    Analysis of Categorical Data
    Instructor(s): Christopher Bilder and Thomas Loughlin

To read about the 2017 section award winners, visit Biometrics Section news.

Celebrate the Significance of Mathematics and Statistics

Sat, 04/01/2017 - 7:00am
April is Mathematics and Statistics Awareness Month


April marks a time to increase the understanding and appreciation of mathematics and statistics. Why? Because both subjects play a significant role in addressing many real-world problems—climate change, disease, sustainability, the data deluge, internet security, and much more. Research in these and other areas is ongoing, revealing new results and applications every day in fields such as medicine, manufacturing, energy, biotechnology, and business. Mathematics and statistics are important drivers of innovation in our technological world, in which new systems and methodologies continue to become more complex.

Be sure to check the ASA’s website throughout April for contests and activities!

“Because of the massive increase in the amount of data available, and because of the important contributions of statistics to making sense of the data, statisticians are in hot demand,” said Ron Wasserstein, the ASA’s executive director. “Jobs in statistics command good pay, have great working conditions, and allow individuals to solve problems that make a difference to the world.”

In the age of Big Data, statistics underlies almost every decision made today, whether it’s the effectiveness of a new drug or treatment or the debut of a mobile device. Statistics is how analysts convert raw data into useful information, from studies of proteins to surveys of galaxies.

Research in statistics and the mathematical sciences is important for its applications and because it trains one in critical thinking and problem solving. From magic squares and Möbius bands to magical card tricks and illusions, mysterious phenomena with elegant “Aha!” explanations have been part of both subjects for centuries.

This month, let’s celebrate mathematics and statistics and the diverse researchers and students in these fields who are contributing so much to furthering discoveries, solving problems, and finding beauty in our world.

Mathematics and Statistics Awareness Month is a program of the Joint Policy Board for Mathematics (JPBM)—a collaborative effort of the American Mathematical Society, the American Statistical Association, the Mathematical Association of America, and the Society for Industrial and Applied Mathematics.


If you would like an 11 x 17 printed copy of the poster mailed to you, email the ASA’s communication manager, Megan Murphy.