The University of Chicago’s Master of Science in Analytics (MScA) is proud to announce as its first faculty director the appointment of Stephen Stigler, Ernest DeWitt Burton Distinguished Service Professor at the Department of Statistics of the University of Chicago. A member of the American Philosophical Society as well as past president of the Institute for Mathematical Statistics, Professor Stigler will deepen the MScA program’s ties to the University while providing oversight to a curriculum focused around preparing students for careers in the dynamic world of data science today.
“My goal will be to work with the program director in making the MScA program the best such program in the city and beyond,” he said. “As a representative of the University I will be looking for new ways to expand and develop the connections between the program and the work we do in the Physical Sciences Division. There is a tremendous amount for each side to learn from the other.”
“I was very impressed by the quality of the program,” Stigler said. “Not all programs go this far into the fundamentals, particularly when it’s a matter of constantly adapting to new analytics trends."
Having had a limited advisory role prior to the MScA program’s creation in 2014, Professor Stigler says he has watched the program grow and adapt its curriculum to the fast-evolving analytics landscape of the past half-decade. In doing so, he also came to appreciate the tremendous value an analytics degree offers graduates today and the strength of the MScA program in particular.
“I was very impressed by the quality of the program,” he said. “Not all programs go this far into the fundamentals, particularly when it’s a matter of constantly adapting to new analytics trends. But the rapid developments in data science mean today’s program might not serve tomorrow’s needs, so the program needs to be constantly aware of change.”
In many ways, Professor Stigler is ideally suited to the role. Having spent his career investigating statistical methods, he stands at once inside and outside the world of analytics and sees both the similarities and differences a professional degree program has with the academic work that has occupied him throughout his career. He also appreciates how each side working together can serve as a recipe to achieve excellence.
“Throughout my career I’ve held a balanced view and I’ve seen how the theoretical and practical sides are in fact very complementary,” he said. “While I put great value in the detached view that strives to understand problems deeply at the fundamental level, I also see engaging with the world as both necessary and valuable. At the end of the day, you’re in a much stronger position if you do both.”
In recent years, Professor Stigler has investigated the history of statistics as well as the key figures who have made decisive contributions to the development of the discipline. In addition to the new understanding he has gained on the material he has taught and researched since arriving at the University in 1979, his engagement with the history of the discipline gave him the tools to respond to what he sees as a shortcoming in many recent popular works of statistics, which boldly claim to make the subject easy.
In 2016, he published The Seven Pillars of Statistical Wisdom with Harvard University Press, a book that, while intended for a popular audience, nevertheless seeks to exemplify the rigor and complexity of statistical thought.
“While I put great value in the detached view that strives to understand problems deeply at the fundamental level, I also see engaging with the world as both necessary and valuable. At the end of the day, you’re in a much stronger position if you do both.”
“Statistical thinking is really hard,” he said. “A lot of the great ideas in statistics look wrong when first encountered, but in fact they’re right. In my book I set forth seven foundational statistical ideas and the complex circumstances from which they arose.”
Taking aggregation as an example—a pillar that is essential to all of today’s data analytics solutions—he points out that arriving at the concept of taking a mean, though it might seem obvious now, was far from intuitive at its origins.
“You lose information when you take an average,” he said. “The individuality of each observation is lost. At the same time, you gain something else. Through looking at the historical development of the idea you come to understand the context that called for that innovation and why it was needed at that point in time.”
In the end, a keen sense for the broader statistical context of the new tools being developed to tackle problems in the constantly shifting world of analytics will be only one of the many important viewpoints Professor Stigler will bring to the MScA program in his role as faculty director. As a longtime member of the University community who has served on any number of outreach programs in the social and physical sciences, his institutional knowledge will also go far to ensuring the mutually beneficial role the MScA program plays within the Physical Sciences Division.