The new IMS President for 2012-2013, Hans R. Künsch, writes:
I took over as president in July at the World Congress, and I have been quite busy since then in my new role, having already sent and received several hundred emails related to IMS. One of the most important duties of the president at the beginning of his/her term is to make appointments to fill all committee vacancies. This made me aware once again how strongly IMS relies on the voluntary services of its members, and I would like to thank on this occasion all those who have contributed their time and knowledge. In close consultation with Past-president Ruth Williams and President-elect Bin Yu, we are currently preparing invitations for new committee members based on the quality of their research, their knowledge of the tasks of the various committees, and their efficiency and reliability—while also making sure that the composition of the committees reflects the diversity of our membership. (The many IMS committees are detailed in the handbook) If you are at the beginning of your career, you might wonder what you can do to be invited as a member of one of these committees one day. Often, a first step is to be asked to join the editorial board of a journal, and for this, editors usually observe who provides timely and informative referee reports in addition to publishing excellent research papers.
Another issue that a president has to deal with is the budget, now finalized for the fiscal year 2013. Fortunately our budget is back in the black after some losses during 2005–2007. So in the near future things are stable and we are able build up our reserves again, but some concerns about the long term financial perspective remain which are mainly due to the uncertainty about how the model of journals subscribed by libraries will develop. I intend to go into more detail on this in one of the coming issues of the Bulletin.
The president is also expected to set an agenda to develop the IMS. So how does my agenda look? Well, my first goal is to build upon the initiatives of my predecessors, as described for instance in the article by Ruth Williams in this same issue last year, like seizing the opportunities of the year 2013 to make statistics and probability more visible, and making sure that the Blackwell lecture can start as planned at the annual meeting in 2014 [Ruth refers to this in her Presidential Address article, on page 6]. One of my special concerns is to promote probabilistic and statistical reasoning in all areas of science. While stochastic methods are well established in some fields like clinical trials, computer vision or finance, in others, such as environmental science, deterministic models are still the dominant paradigm since they are based on laws of nature and our mechanistic understanding of the processes in the system. However, there are also substantial uncertainties about parameters, inputs or unresolved processes. Recently the numerical analysis community has started to treat such uncertainties as stochastic so that the solutions of the associated partial differential equations also become random. They are developing faster alternatives to Monte Carlo in order to determine the distribution of these solutions (see, for example, A. Cohen, R. Devore and C. Schwab, “Analytic regularity and polynomial approximation of parametric and stochastic elliptic PDE’s,” Analysis and Applications, Vol. 9, (2011), 11–47). I therefore think that probabilists and statisticians on the one hand, and numerical analysts and applied mathematicians on the other, can profit a great deal from closer cooperations. To this end, I am glad that IMS has recently become an associate member of ICIAM, the International Council for Industrial and Applied Mathematics. I will do my best to ensure that we do not miss the opportunities this offers.
Summer is also the traveling season, and I attended three conferences: the ISBA World Meeting in Kyoto, the Asia-Pacific Rim Meeting in Tsukuba, and the World Congress in Istanbul. At all three occasions, I was impressed by the diversity and quality of the lectures I attended. I particularly enjoyed listening to lectures which gave me an introduction and overview on a topic where I am not actively working. Getting a glimpse into the the beauty and the power of mathematical ideas is one of the attractions of our profession! Although the meetings I attended were small compared to the JSM, sometimes the program with a large number of interesting parallel sessions made the choice difficult for me. I believe that we should consider seriously the possibility of having fewer oral presentations and more posters in big meetings. Currently, most of us prefer to give an oral presentation than a poster, but if the poster sessions are well organized (at convenient times, without competition from lectures, and combined with some social event), then they can be more rewarding both for presenters and the audience, as the experience from other events like the conferences on Neural Information Processing Systems (NIPS) or the meetings of the American or European Geophysical Unions show. Apparently, the program committee for the next JSM is considering an experiment having poster sessions combined with five-minute talks to advertise the poster which seems an interesting and worthwhile idea to me.
A key concept and challenge which was often mentioned at these conferences was “Big Data” or the “Data Deluge”. Undoubtedly, the amount of data that is collected every day has reached an unbelievable level, in particular through our use of services on the web. However, datasets that are too big for most conventional statistical analyses have been around for more than 20 years. I think it is instructive to look also back at what people have said about this topic, e.g. Peter J. Huber in “Massive Datasets Workshop: Four Years After,” Journal of Computational and Graphical Statistics, 8 (1999), 635–652, in order to see what has changed since and which of the challenges are still with us.
During the long hours on the plane and some quiet times in between conferences, I also had the pleasure to read books. One of them, Thinking, Fast and Slow by Daniel Kahneman (Penguin Books 2012), has many connections to probability and statistics. In particular, it has several astonishing examples of how our mind fails to handle statistical information in a coherent and logical way. Some of these will make a welcome addition to the discussions in my elementary statistics courses.
I hope that these varied comments have given you an impression of me as incoming IMS president. I appreciate your continued support of IMS very much, and if you have any ideas about how IMS could fulfill its mission even better, please contact me at email@example.com.