Our second roundtable discussion took place after Pantelis Loupos’ Statistics seminar talk on “Starting Cold: The Power of Social Networks in Predicting Non-Contractual Customer Behavior". His talk slides can be found here.

The following note summarizes this roundtable discussion.

The topics was Data science needs and issues in Business Schools, e.g.,

• How the students in business school/school of management are trained in data science/machine learning now? Are they also taking some of the courses offered by our 4 departments (CS/ECE/Math/Stat)?

• What opportunities (problems, projects) in data science/machine learning could the business school provide us in CS/ECE/Math/Stat faculty, postdocs, and graduate students?

• How about potential collaborations between the graduate school of management and our 4 departments?

These are timely topics because the US will experience a shortage of 1.5 million managers and analysts who can use big data to make effective decisions this year, according to McKinsey & Company, and consequently, some business schools such as London’s Imperial College Business School, have taken a novel approach to data science education by launching their Data Observatory (see, e.g., the article by BusinessBecause for the details).

The majority of this roundtable was devoted to the discussion on data science related activities (both research and teaching) of the UC Davis Graduate School of Management (GSM), of which the speaker is a member. He invited Professor Ashwin Aravindakshan, who is the academic director for the Masters of Science in Business Analytics (MSBA), who gave an overview of the role of data science in the GSM. He first described the MSBA degree progam, which is taught in San Francisco, and consists of classroom courses as well as collaborative, hands-on business analytics projects working with industry partners. This one-year degree program is heavily focused on business-oriented applications; their emphasis is more on how to apply the state-of-the-art models to interesting and challenging business-oriented applications rather than developing new models for those applications.

More broadly, there are about 8 faculty members (among 30 or so in the GSM) who work on data science research, such as both Professor Loupos and Professor Aravindakshan. The GSM does not have a Ph.D. degree program presently and only recently added a post-doc position. The majority of their advisees come from one of the 4 disciplines represented in the UCD4IDS program. Their master students have too compressed a schedule to be able to take the background coursework necessary to do research work in data science, so they are very much interested in collaborating with graduate students from mathematics, statistics, and computer science who have that rigorous background, but still need to gain an understanding of the peculiarities of business problems and how to frame their analysis in actionable terms. An attendee suggested that they consider adding a designation, in a similar manner to the health analytics certificate.

[Scribe: David Weber (GGAM)]