A0750
Title: Optimal designs for sparse functional data
Authors: MingHung Kao - Arizona State University (United States) [presenting]
Abstract: Sparse functional data analysis (FDA) is powerful for making inferences on the underlying random function when noisy observations are collected at sparse time points. Knowledge of optimal designs that allow the experimenters to collect informative, functional data is crucial for a precise inference. We propose a framework for selecting optimal designs to precisely predict functional principal and empirical component scores. A relevant generalization of previous results on the design for predicting individual response curves is given. Optimal designs are obtained, and evaluate the performance of commonly used designs. It is demonstrated that without a judiciously selected design, statistical efficiency can be lost.
