I am soon to finish my term as chair of the Department of Statistics (this summer July 1st), turning the reins over to Deb Nolan, for her second tour of duty.

I've quite enjoyed being chair of the department in this era. Our department has been growing, according to many of the metrics that matter most — the number of students in our major (which is arriving at 500), the level of contact with industry, the amount of attention being paid to our discipline by other areas of academia, and the range and scope of our applied work.

We are also slowly but surely growing our faculty, with four strong new hires over the past three years — Nike Sun, Will Fithian, Peng Ding and Sam Pimentel. Nike is a leading young probabilist (e.g., she has just been named the 2017 winner of the Rollo Davidson Prize), and Will, Peng and Sam are major talents in statistics, spanning applied and theoretical research. All three have strong interests in causal inference, and their arrival has restored Berkeley's place as a leading center of causal inference (a turn of events that would have pleased Jerzy Neyman).

A major driving trend in the field of statistics over the past decade has been the rise of "data science." This term has become widely used in academia, industry and government to refer to a blend of the computational sciences and the inferential sciences. Data science focuses on the opportunities and challenges engendered by the ubiquity of data collection and computing infrastructure in modern life, the necessity of bringing statistical perspectives to bear in the collection, analysis and interpretation of these new data streams, the need to combine computational and statistical perspectives to develop scalable, statistically-sound methodology, and the critical need to take into account the implications of data science for trustable infrastructure, replicable science and data-aware public policy.

We have taken a leading role in the definition and the building of data science at Berkeley. A notable achievement has been the development — in collaboration with Computer Science — of an innovative new freshman-level course on data science (DS8) that combines three perspectives: "inferential thinking, computational thinking, and real-world relevance." (See data8.org). The course has grown rapidly in its three instantations to serve several hundred freshman per semester, and its enrollment will likely top one thousand this coming fall. We have also developed four new courses at the sophomore and junior levels that blend inference and computing, and several more are in the works.

We have also participated in the ambitious process of defining new organizational structures at Berkeley to accommodate data science and to further align the computing sciences, the statistical sciences, and the large number of disciplines that are being impacted by the ubiquity of data and computing and inferential problems. This process has begun to yield fruit. The campus has announced the creation of a new "Division" that is to hold several existing Berkeley departments in these areas, including Statistics, as well as some new departments. A new interim Dean is soon to be announced who will lead the Division. We look forward to forging new ties on campus in the context of this new Division, and we have high hopes for the long-term impact of this innovative new alignment within academia and in government and industry. It does not seem to be too much of a stretch to view the alignment of "inferential thinking, computational thinking, and real-world relevance" as akin to the emergence of the liberal arts in centuries past.

(Photo by Peg Skorpinski)