Online Program Home
My Program

Abstract Details

Activity Number: 414 - Models for Environmental Processes
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics and the Environment
Abstract #305172
Title: Robust Functional Multivariate Analysis OfVariance with Environmental Applications
Author(s): Zhuo Qu* and Marc Genton and Wenlin Dai
Companies: KAUST and King Abdullah University of Science and Technology and Renmin University of China
Keywords: Functional data; Image data; Median Polish; Multivariate analysis of variance; Robustness; Spatio-temporal data

This paper proposes median polish for functional multivariate analysis of variance (FMANOVA). As an alternative to classical mean estimation, functional median polish estimates the functional grand effect and factor effects on the basis of functional medians in one-way and two-way additive FMANOVA models. Functional medians are computed according to the simplicial band depth for multivariate functional data. The corresponding parametric and nonparametric tests are generalized to evaluate if the functional medians in various levels of the factors are the same. Functional factors are visualized with functional boxplots or heatmaps depending on whether thefunctional data are curves or images, respectively. Simulation studies illustrate the robustness of our functional median polish in various scenarios compared with the results from classical FMANOVA fitted by means. Three environmental datasets are considered to demonstrate thatour median polish is robust against outliers in practical implementations.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program