A1017
Title: Sufficient dimension reduction for conditional quantiles for functional data
Authors: Eliana Christou - University of North Carolina at Charlotte (United States) [presenting]
Eftychia Solea - Queen Mary University of London (United Kingdom)
Shanshan Wang - University of North Carolina at Charlotte (United States)
Jun Song - Korea University (Korea, South)
Abstract: Functional data analysis is an important research area with the potential to transform numerous fields. However, existing work predominantly relies on the more traditional mean regression methods, with surprisingly limited research focusing on quantile regression. Furthermore, the infinite-dimensional nature of the functional predictors necessitates the use of dimension-reduction techniques. Therefore, this gap is addressed by developing dimension-reduction techniques for the conditional quantiles of functional data. The convergence rates of the proposed estimators are derived, and their finite sample performance is demonstrated using simulation examples and a real dataset from fMRI studies.
