This $300,000 Project Grant award from the National Science Foundation (NSF) Division of Mathematical Sciences (CFDA 47.049 - Mathematical and Physical Sciences) supports research to develop new nonparametric theory and methods for analyzing censored data. The key objectives are to:
-
Establish a distributional theory for spline-based estimates in various censored data models, which is currently lacking in the literature.
-
Investigate the use of deep neural networks with full likelihood-based loss functions to estimate conditional distributions and predictions for censored survival data, including cases with complex, time-varying predictors.
This work aims to provide robust, flexible statistical tools for biomedical, epidemiological, and other scientific studies where survival times are often censored. The project will foster interdisciplinary collaborations and contribute significant methodological advancements with versatile real-world applications. No sub-awards are planned under this grant, which is scheduled to run from October 2024 through September 2027.
Generated 1/28/25, 7:18 AM