Functional data

Classification of Functional Data by Detecting the Discrepancy of Second Moment Structure of Scaled functions

-

KAUST

This talk presents a new classification method for functional data. We consider the case where different groups of functions have similar means so that it is difficult to classify them based on only the mean function. To overcome this limitation, we propose the second moment based functional classifier (SMFC). Here, we demonstrate that the new method is sensitive to divergence in the second moment structure and thus produces lower rate of misclassification compared to other competitor methods. Our method uses the Hilbert-Schmidt norm to measure the divergence of second moment structure. One important innovation of our classification procedure lies in the dimension reduction step. The method data-adaptively discovers the basis functions that best capture the discrepancy between the second moment structures of the groups, rather than uses the functional principal component of each individual group, and good performance can be achieved as unnecessary variability is removed so that the classification accuracy is improved. Consistency properties of the classification procedure and the relevant estimators are established. Simulation study and real data analysis on phoneme and rat brain activity trajectories empirically validate the superiority of the proposed method.

Zhuo Qu

Ph.D. Student, Statistics
Zhuo Qu is a PhD candidate in statistics, under Prof. Marc Genton. Zhuo likes to use statistics methods to interpret the data in the applications and find the new possibility of exploring the new statistics model. Being eager to learn and experience the uncertainty in the research and life, she also like chatting, swimming and piano. Her goal is to be a spare-time pianist among professional statisticians. Education and Early Career King Abdullah University of Science and Technology, Thuwal, January 2019 - December 2022, PhD in Statistics King Abdullah University of Science and Technology

Jonatan A. González

Postdoctoral Research Fellow, Statistics
Jonatan's research interests lie between spatial statistics and coding, emphasising point processes, epidemiology, infectious disease modelling and, generally, application-focused methods. He has a bachelor's degree and a master's background in mathematics. He completed his PhD in spatial and spatio-temporal point processes at the University Jaume I. Jonatan's research projects include techniques for dealing with point patterns: replicated observations, spatio-temporal analysis, graphical methods, hypothesis testing, experimental design and local characteristics. In addition to the core