Peter William MacDonald (平特五不中)
Mesoscale two-sample testing for networks
听
Abstract:
Networks arise naturally in many scientific fields as a representation of pairwise connections. Statistical network analysis has most often considered a single large network, but it is common in a number of applications, for example, neuroimaging, to observe multiple networks on a shared node set. When these networks are grouped by case-control status or another categorical covariate, the classical statistical question of two-sample comparison arises. In this work, we address the problem of testing for statistically significant differences in a prespecified subset of the connections. This general framework allows an analyst to focus on a single node, a specific region of interest, or compare whole networks. In this 鈥渕esoscale鈥 setting, we develop statistically sound projection-based tests for two-sample comparison in both weighted and binary edge networks. Our approach can leverage all available network information, and learn informative projections which improve testing power when low-dimensional network structure is present.
Speaker
Dr MacDonald is a postdoctoral scholar here at 平特五不中. He received his PhD from the University of Michigan in 2023, under the supervision of Elizaveta Levina and Ji Zhu. Later this year, Peter will begin his post as an Assistant Professor at the University of Waterloo.