The Department of Statistics at the University of Michigan is pleased to announce a new NSF grant, "Understanding dynamic big data with complex structure”.  The funding of $2.5M over the next 5 years will support 2-3 postdoctoral researchers, 5-6 PhD students, and 5-6 undergraduate research assistants per year.  This award is through the Research Training Groups program (RTG), part of the Workforce Program in the Mathematical Sciences and is aimed at increasing the domestic STEM workforce and participation of group underrepresented in STEM fields.  Liza Levina serves as PI; Xuming He, Edward Ionides, and Susan Murphy are co-PIs, and seven more Statistics faculty members are affiliated with the group:  Moulinath Banerjee, Johann Gagnon-Bartsch, Brenda Gunderson, Ben Hansen, Stilian Stoev, Ambuj Tewari, and Ji Zhu.  

The goal of this grant is to train undergraduate and graduate students and postdocs in modern research and teaching methods in statistics, preparing them for the new unique challenges brought on by the emerging era of big data.  For the new types of statistical problems we now aim to solve, the size of available data has grown immensely in many cases, and the nature of the data has changed no less dramatically.  We now work routinely with data that combine many different kinds of observations, from genetic data to brain images to smartphone data. This creates a need for new training approaches and their close integration with current research directions.  It also creates an opportunity for recruiting undergraduates into the field, increasing and diversifying the domestic STEM workforce.

The research core of this project is organized around three interlinked streams:  statistical network analysis, inference for dynamic systems, and sequential decision making.  Each research stream will offer a short intensive graduate course, and a regular interdisciplinary student workshop.    Equally importantly, the streams will collaborate on topics that cut across these areas, such as inference for dynamically evolving networks or the role of social connections in predicting behavior and their impact on sequential decision making.  All research streams have broad applications to areas beyond statistics, such as neuroimaging, infectious disease transmission, and mobile health interventions, and all trainees will be encouraged to pursue interdisciplinary collaborations outside the group, including through MIDAS, the new Michigan Institute for Data Science.