Federal Project Grant Award Summary

The University of California, Berkeley received a $180,000 Project Grant from the National Science Foundation's Division of Mathematical Sciences under the Mathematical and Physical Sciences program (CFDA 47.049), awarded July 15, 2026, with completion targeted for June 30, 2029. This collaborative research project develops novel statistical and machine learning methods for analyzing multivariate multi-replication point process data—specifically, data that records when events occur over time. The research addresses critical analytical gaps in handling irregular temporal event data across multiple event types and large numbers of subjects, with direct applications to electronic health records and neuroscience research.

The project delivers three complementary classes of statistical models: generalized additive models for examining associations between scalar responses and high-dimensional point process predictors; graphical models for estimating dependence networks among multiple point processes through latent random intensity functions; and structural equation models for learning directional influences among events. Beyond methodological development, the award supports workforce development and broader scientific dissemination through research-based training, mentoring, outreach activities, and freely available software resources. The research advances the mathematical and computational foundations of event-time data analysis with potential to improve clinical knowledge discovery from electronic health records and enhance understanding of neuronal activity and brain function.