On November 3-4, Winona State statistics and data science students participated in the fantastic MinneMUDAC 2017 data analytics competition. Students worked in teams of up to five students for one month analyzing de-identified administrative medical and pharmacy claims data provided by Optum. The competition required students to analyze complicated data on health insurance claims made by Type-II diabetics. The event was hosted by MinneAnalytics at the Optum campus in Bloomington, Minnesota.
The event began the evening of November 3. MinneAnalytics provided dinner and a professional Q&A panel of data scientists. The event kicked off in earnest on Saturday, November 4. Students gave their 5-minute presentations to teams of judges, and were scored on a variety of criteria ranging from analytic acumen; presentation organization; and team synergy. Some weight was also given by how accurately teams could predict the top-6000 most expensive Type-II diabetics for the next calendar year for a held-out target data set of patients for which the actual insurance costs were known only to the event organizers.
The top five teams from the first round went on to present to all judges in a final round. From this final round, teams were given awards for “Best overall”; “Analytic acumen”; and “Serendipitous discovery.”
There were 22 total undergraduate teams who participated. We were very excited that of the top five teams to proceed to the final round, three were from Winona State! Of these three, one won “Best overall” and another the award for “Analytic acumen.” We were very proud of all our students who participated. Each one of them took time out of their busy semesters to gain fantastic professional development experience. From real-life data analysis of a very messy and complicated data set to presenting their results to panels of industry and academic professionals, it will be an experience that will serve them all very well.