IMS President Richard Davis writes:
During my one-year apprenticeship as President-elect of IMS, I was often queried about my agenda for IMS. The word “agenda” can conjure up both positive and negative feelings. If a dean enthusiastically espouses a new agenda, then faculty are apt to cringe at the idea of change. So instead of talking about an agenda, let me take a stab at how I would like to guide IMS going forward.
Since the term of an IMS President lasts only one year (I am not lobbying for more!), major changes typically take place over more than one presidency. With this in mind, I intend to build on the initiatives that former presidents Bin Yu and Erwin Bolthausen started and begin some of my own.
In her 2014 presidential address in Sydney, Bin suggested that statisticians should own data science. This sounded great even if one does not know exactly what data science (DS) entails. Who would not want to own anything with data in the title? I will leave it to others to provide a clear description of DS—for now I’ll stick with the “I know it when I see it” definition. Unlike previous “next big things”, which tend to have a short shelf life, data science appears to have staying power. This can be seen at many universities, including my own, which are investing huge sums of money in creating Institutes and Centers of Data Science. Statistics needs to play a central, if not a leading role in these developments. The IMS council has just approved a new Data Science group, headed by David Dunson, and has discussed other ideas about capturing more of the statistics/probability components, widely defined, of data science.
In his presidential address at JSM, Erwin discussed “relations between statistics and probability theory.” He expressed concerns about these two fields drifting further apart, to the detriment of both. IMS is one of the few professional societies that has large representations of both statisticians and probabilists. I have heard anecdotal remarks that probabilists are more aligned with and supportive of mathematical societies than IMS. For certain branches of probability, mathematical societies provide a more natural fit. Even though IMS publishes some of the leading journals in probability theory, we must do more to attract probabilists, especially young researchers. Since access to papers published in IMS journals is rather simple to obtain electronically, it is no longer necessary to be a member to possess your own copy of the Annals. The argument for being an IMS member has to go beyond just producing superb journals, although this is a vital service to the entire profession. We should all be committed to a society that promotes and advances our field, provides pathways for reviewing and disseminating knowledge, and facilitates interactions and collaborations. While IMS is not perfect, we need to make the case, especially to new researchers, that IMS is a worthwhile investment on their careers.
IMS has supported the New Researchers Conference that precedes JSM for a number of years now. We now have a New Researchers Group, spearheaded by Alex Volfosky. The website, which I expect will be dynamic and informative, will be coming online soon. While the NR group will be involved in a number of activities, including organizing the New Researchers Conference, sponsoring sessions at IMS annual meetings and other IMS co-sponsored conferences, and organizing workshops, its main goal is to provide engagement and involvement of early career researchers within IMS. The formation of this group is an exciting development, which is long overdue.
Although there are many issues, both large and small, that we are trying to address in IMS, my intent is to create opportunities for members to connect more closely with the society. IMS has a mechanism for creating groups, which essentially consists of individuals coming together along a common interest, and calling themselves a group. The FPS (Finance: Probability and Statistics) group has been reasonably successful with organizing annual meetings. So far, IMS has not provided much support for these groups—most do not even have a webpage, and even if they do, they are essentially invisible from the IMS website. I would like to change the model for groups, and the two new ones (New Researchers and Data Science) provide a template for the formation of new groups. In the last several years, IMS has opened up its process for nominating individuals to named lecturers and for proposing sessions at meetings. The formation of groups is essentially a continuation of providing members more direct access to these opportunities. The bottom line is to provide a mechanism for individuals to have more of a voice in a large organization such as IMS. With data science flexing some muscle, IMS has to be forward thinking in trying to engage this group. Already discussions are underway about organizing jointly sponsored sessions on DS with ASA and IEEE at our annual meetings. This is just a start.
Of course, your suggestions and comments for improving IMS are always welcome. I am committed to ensuring that IMS remains a vibrant society and responsive to its members while looking to the future.