A0950
Title: A non-smoothing framework for inference on functional means
Authors: Hsin-wen Chang - Academia Sinica (Taiwan) [presenting]
Abstract: A nonparametric inference framework that applies to occupation time curves derived from wearable device data is introduced. Motivated by the right continuity of these curves, a non-smoothing approach is developed that involves weaker conditions than existing conditions imposed when using smoothing to estimate functional means under a fixed dense design. Notably, the procedure allows discontinuities in the functional covariances while accommodating the discretization of the observed trajectories. Under this non-smoothing framework, an empirical likelihood method is devised to construct confidence bands for the functional means. The method utilizes the known optimality of empirical likelihood. It also respects range and monotonicity constraints on occupation time curves. A simulation study shows that the proposed procedures outperform competing functional data procedures.
