Federal Project Grant Award Summary

The University of North Carolina at Chapel Hill received a $100,000 Project Grant from the National Science Foundation's Division of Mathematical Sciences (CFDA 47.049) awarded June 1, 2026, with a completion date of May 31, 2029. This collaborative research initiative develops statistical theory and methodology for quantifying rates of change and gradients in spatiotemporal datasets, with primary application to biomedical and neuroimaging research. The project delivers software tools and computational methods for identifying and assessing boundaries that track significant changes across large, complex spatiotemporally indexed datasets, leveraging low-rank projection-based approximations to Gaussian processes and scalable Bayesian factor models.

Key deliverables include rigorous statistical inference techniques for rates of change associated with predictive and graphical predictive processes, posterior inference probability distributions within Bayesian frameworks, and methodology for tracking rapid directional changes across space-time smooth surfaces. The project incorporates research training opportunities for graduate students and addresses substantive scientific questions in biomedical and neuroimaging domains where detecting regions of rapid spatial and temporal change is critical. Methodological developments align closely with machine learning and artificial intelligence applications, positioning the work to advance statistical inference capabilities for massive, multivariate spatiotemporal data analysis across multiple scientific disciplines.