Title: Simultaneous inference in functional data analysis
Authors: David Degras - University of Massachusetts Boston (United States) [presenting]
Abstract: Statistical methods will be discussed for simultaneously inferring functional parameters such as mean, covariance, and regression functions. From a theoretical perspective, we will start with the standard case of i.i.d., densely observed functional data (FD) and then discuss challenges posed by other observation schemes (e.g. sparse or partially observed FD) and dependence mechanisms (e.g. functional time series, spatial FD). A numerical study will provide insights into the computational and statistical performances of simultaneous confidence bands methods. Finally, we will consider recent lines of FD research, such as manifolds and partially observed FD.