Published:

A doctoral student in the Department of Applied Mathematics and Statistics took top honors for his presentation at the American Statistical Association’s recent Nonparametric Statistics Student Competition.

Joshua Agterberg, whose work focuses on theoretical problems in data science and statistics, won for his presentation on “Nonparametric Two-Sample Hypothesis Testing for Random Graphs with Negative and Repeated Eigenvalues,” which tackles the problem of determining whether two networks (with different vertices) have the same distribution. Co-authors on the paper were Minh Tang of North Carolina State University and Carey Priebe, a professor of applied mathematics and statistics and Agterberg’s adviser.

Joshua Agterberg

“An example where this would be useful is determining whether two social networks have the same connectivity properties: for example, if you have snapshots of two different Facebook friendship networks (with maybe different people in each network) at different times, you may want to determine if these two different networks have the same types of connectivity,” said the Milwaukee native, who earned his bachelor’s degree at the University of Wisconsin-Madison.

The presentation was so successful, Agterberg believes, because he not only geared it to a general statistics audience, but also focused on how the problem in question is different from related classical statistics problems.

“I tried to convey how our theoretical results overcome these difficulties without going into too much detail. I wanted to keep my presentation short and sweet, so I mostly just emphasized how certain classical assumptions break down for this problem,” said Agterberg. “And while our paper is primarily theoretical, our procedure can more or less be used ‘out of the box’ for generic network data, making it readily applicable to essentially any scientific or engineering problem that has to do with comparing two networks.”

The “Best Presentation” honor comes in the wake of another: Agterberg also just learned that he won the 2021 Institute of Mathematical Statistics’ Hannan Graduate Student Travel Award, which provides money for travel to any IMS-sponsored or co-sponsored meeting.

Agterberg is an AMS Teaching Fellow currently instructing a one-credit course, “Statistics and Data Science with Networks,” for first- and second-year engineering students.  This research was also supported in part by a fellowship from the Mathematical Institute of Data Science (MINDS).