Sponsor/Partner: NSF National Science Foundation
Project Description/Goals: This project aims to develop a regularized halfspace depth for functional data, providing the first solution to a long-existing problem of degeneracy. Building on this foundation, the next phase will focus on extending the method to account for uncertainties in sparse and noisy longitudinal observations, expanding the few depth notions designed for this type of object. Additionally, a restricted metric halfspace depth will be explored to enable the detection of shape outliers with distinct features. The project will also propose practical graphical tools for outlier detection.
CTML Faculty Involved: Alejandro Schuler Ph.D.
