One of the final acts of the outgoing IMS President is to deliver the Address at the IMS annual meeting. Ruth Williams, who is now Past-President, delivered this at the World Congress in July.
The IMS mission is “to foster the development and dissemination of the theory and applications of statistics and probability”. Next year, 2013, we have a unique opportunity to heighten the profile of probability and statistics, and I would like to ask all of you for your help in capitalizing on this opportunity.
The year 2013 is a double feature year: it is the International Year of Statistics and the year of celebration of the Mathematics of Planet Earth. The year marks the anniversaries of a number of historic publications, including those of Jakob Bernoulli and Thomas Bayes. The celebrations are intended to be broadly inclusive and to feature a wide array of topics in probability and statistics.
In cooperation with other mathematics and statistics societies, the IMS is already planning a number of activities associated with one or both of these themes.
For example, at next year’s Joint Statistical Meetings, in cooperation with the Bernoulli Society, to celebrate the 300th anniversary of the publication of Jakob Bernoulli’s Ars Conjectandi, we will be cosponsoring a public lecture by David Spiegelhalter, who is the Winton Professor of the Public Understanding of Risk at Cambridge University.
Invited special sessions relating to the role of probability and statistics in planet earth are planned for major meetings. For example, there will be a session organized by Tilmann Gneiting on spatial and space time statistics at the European Meeting of Statisticians in July 2013. Special issues for some of our journals are also in the planning stages.
There is room for much more, especially, at the grass roots or individual level. The themes of 2013 provide us all with a unique chance to highlight the past and present achievements in statistics and probability and to ponder possible future prospects.
Within our profession, the IMS is well known for its activities related to research in probability and statistics, including our publication of high quality journals at reasonable cost, and sponsorship or cosponsorship of lectures and conferences featuring the latest research developments in our fields. The year 2013 provides us with an opportunity to further these efforts, but also to make our achievements more broadly known.
I will illustrate some initiatives that are already under way and outline ways in which you as individual members can help this effort; although I encourage you all to think of other ways in which you might contribute. My description will fall under three headings, roughly corresponding to the past, present and future of the IMS.
In 2010, the IMS celebrated its 75th anniversary. Regrettably, we have now seen the passing of many of the great contributors to the early years of our society and to the foundations of modern Probability and Statistics. In an effort to create a lasting resource highlighting our rich history, last year, the IMS created the position of Scientific Legacy Editor, with Paul Shaman being appointed as the first such editor. Paul is involved in developing webpages that feature the many contributions made to probability and statistics (and their applications) by all of our distinguished members. It is anticipated that this will be a valuable resource for increasing the profile to the world at large, as well as to our members, of our scientific heritage.
As a first step, Paul Shaman, with the technical assistance of Jim Pitman, has already made significant progress on establishing a reliable data set for our IMS Fellows. This project is ongoing and to support these and related efforts, the IMS has established a Scientific Legacy fund to which contributions are welcome. Stay tuned for the unveiling of the revamped IMS Fellows page.
The term Scientific Legacy as I have used it here is meant to be broadly construed. I believe that through this project we can develop a rich set of web resources highlighting past and present research achievements in probability and statistics.
Complementary to these web-based efforts, the IMS has a number of named lectures and awards honoring key contributors to probability and statistics. Last year, in cooperation with the Bernoulli Society, the IMS established a joint lecture in probability and stochastic processes, to be titled the Schramm lecture. The first Schramm lecture will be delivered by Itai Benjamini in 2013 at the Stochastic Processes and their Applications meeting in Boulder, Colorado. The expenses of the lecturer will be supported in part by the Schramm lecture fund. I am pleased to acknowledge the generous initial contributions to this fund by Microsoft Research and Cambridge University Press, as well as several IMS members; of course, contributions continue to be welcome. More recently, the IMS has entered the final stages in establishing a Blackwell lecture to honor the memory of David Blackwell.
We certainly live in a very exciting time for our fields: not only are there challenging foundational problems, there is a seemingly ever- expanding list of applications in a broad range of fields in need of new research developments in probability and statistics.
Now, the IMS’s high quality journals and sponsored/cosponsored conferences, such as the World Congress in Probability and Statistics that we are currently attending, do an excellent job of conveying the most recent research developments in probability and statistics to our own community. And high profile awards—recent Fields Medals to researchers in probability and related fields, the Gauss Prize awarded to K. Itô and the Abel Prize awarded to S.R.S. Varadhan, the U.S. National Medals of Science awarded to Brad Efron and S. R. S. Varadhan—all help to draw broader attention to the outstanding research achievements of our field.
While one might be tempted to rely on the wisdom of George Bernard Shaw (“It is the mark of a truly intelligent person to be moved by statistics”) and to leave our efforts at that, I believe that we need to be more proactive in illustrating the research achievements of probability and statistics to those in other areas of science and engineering, and to the public more generally.
Now of course, it is challenging to do this and perhaps the quote of Richard Feynman comes to mind: “If I could explain it to the average person, I wouldn’t have been worth the Nobel Prize” (People, 1985). However, I believe that it is imperative that we make a special effort to highlight the importance of supporting research in probability and statistics, and to emphasize the need to involve experts in interdisciplinary work involving stochastics and statistics, and in the teaching of probability and statistics at all levels. To emphasize this point, I will give one recent example to indicate why I believe a proactive stance is required. Although this example comes from the United States, there are signs of similar developments in some other countries.
Recently, a report by the Council of Advisors on Science and Technology to the President of the United States recognized inadequate pre-college mathematics preparation as a bottleneck in undergraduate STEM (Science, Technology, Engineering and Mathematics) education. As part of a proposed solution, the Committee recommended teaching and curriculum development of University level mathematics by faculty from mathematics-intensive disciplines other than mathematics, including physics, engineering, and computer science; and furthermore, for a new pipeline to be opened for producing elementary and high school mathematics teachers from undergraduate and graduate programs in mathematics-intensive fields other than mathematics.
This recommendation was part of an initiative to support the worthy aim of increasing the number of university graduates with degrees in science, technology, engineering or mathematics. Perhaps the recommendation is not so surprising given that the committee that made it, did not include a single professor of mathematics or statistics, although it did include professors of biology, computer science, chemistry, economics, engineering, geology and physics, presidents of universities and representatives of scientific research institutes and industry. (You can read the report here).
After the initial shock of hearing this recommendation, one might reflect on how history appears to repeat itself: in 1940, Harold Hotelling, in an address to the IMS (Annals of Mathematical Statistics, 11 (1940), 457–470), argued for the teaching of the fundamentals of university-level statistics by specialists in statistics, rather than those involved only in applications thereof.
Taking an optimistic view, I believe this should serve as a wake-up call that there is an urgent need for us to convey to those outside of our field the importance of supporting fundamental research in probability and statistics, and in involving experts in probability and statistics in interdisciplinary research and in the teaching of these subjects. I believe that these efforts will be all the more effective if we combine with other mathematics and statistics societies in this effort.
While we already have strong cooperation with several other societies that promote probability and statistics research, such as the American Statistical Association, the Applied Probability Society of Informs, the Bernoulli Society, and the ISI, the IMS continues to reach out to other societies, especially those using mathematical and statistical methods in interdisciplinary research.
As part of this effort, recently the IMS joined as an associate member of ICIAM, the International Council for Industrial and Applied Mathematics, to foster stronger ties to the applied mathematics community. ICIAM holds a quadrennial congress featuring the latest developments in industrial and applied mathematics and increasingly features topics related to stochastics and statistics.
Amongst the celebrations of 2013, there will be various opportunities to ponder the future of probability and statistics and I encourage you all to participate in these activities.
One area in which I expect there will be continuing growth is in problems spurred by applications that require expertise from several different fields and which may lie at the intersection of the areas of mathematics, probability, statistics, informational and computational sciences.
While I am sure there are many different directions that will be contemplated, I would like to mention two here, as examples of some of the prospects for the future.
The first relates to quantitative biology, which is sometimes called the new biology. While some parts of biology such as ecology and evolution, neuroscience and genomics, have developed sophisticated uses of probability and statistics in collaboration with experts in these fields, some other parts of biology have not previously sought such expertise intensively. However, there seems to be a sea-change occurring in biology, where there is a desire to have a blend of experiments, modeling and analysis, and computation, in many areas of biology. To make the connections between models and experiments will often require a knowledge of mathematics, stochastic processes, statistics and computational methods, as well as biology, of course. While this is a natural place to involve probabilists and statisticians, I think it will require a special effort to make connections, as many biologists are not highly trained in mathematics or statistics and may not think to seek out experts in probability and statistics to help them. I also believe that this is an area where collaborations of teams from many different quantitative areas will be very beneficial and hence the need for us to connect with applied mathematicians, physicists, computer scientists, engineers, and so on.
The second area that I would like to mention is networks. Now, networks of various kinds have been studied for some years and we have all heard of newer social networks, but there appears to be a growth in problems associated with complex networks where information can play an important role and can influence outcomes.
An example is transportation systems such as freeways where the availability of real time traffic information and economic incentives can alter the behavior of individuals and thereby affect quantities such as congestion in a network. Such systems are sometimes called societal networks. The interplay of economics, sociology, probability and statistics here provides for challenging problems.
Of course, there is a wealth of other topics of growing interest, I simply chose two that are colored by my own experience. Indeed, one of the strengths of probability and statistics is that there are so many connections to other fields and to applications. This also presents a challenge in that we do not have one identity, but a whole spectrum of appearances.
In closing, let me emphasize that I believe 2013 provides us with a unique opportunity to feature the achievements and potential contributions of probability and statistics to those outside of our field, including researchers in other fields, teachers, the general public, and government decision makers.
To be effective, we all need to try to help in whatever ways we can. Whether it be by sponsoring public lectures at your University, or organizing events for undergraduate majors, or writing short accessible summary snapshots of research (“nuggets”), or contributing survey articles to one of our survey journals, or inviting your government representative to events highlighting the importance of probability and statistics in interdisciplinary research and the teaching of these subjects, or contributing to brainstorming sessions on future new directions, or other activities you might think of, I call on you all to seize the opportunity afforded by this special year devoted to highlighting probability and statistics.