Assessing extrema of empirical principal component functions

DOI10.1214/009053606000000371zbMATH Open1113.62074OpenAlexW4298166670MaRDI QIDQ2500461FDOQ2500461

Authors Céline Vial, Peter Hall

Publication date 24 August 2006

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0608025

zbMATH Keywords

bootstrapfunctional data analysismodeprincipal components analysisturning pointperturbationsubsamplebootstrap likelihoodKarhunen-Loeve representationshoulder point

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Nonparametric statistical resampling methods (62G09) Statistics of extreme values; tail inference (62G32)

Abstract: The difficulties of estimating and representing the distributions of functional data mean that principal component methods play a substantially greater role in functional data analysis than in more conventional finite-dimensional settings. Local maxima and minima in principal component functions are of direct importance; they indicate places in the domain of a random function where influence on the function value tends to be relatively strong but of opposite sign. We explore statistical properties of the relationship between extrema of empirical principal component functions, and their counterparts for the true principal component functions. It is shown that empirical principal component funcions have relatively little trouble capturing conventional extrema, but can experience difficulty distinguishing a ``shoulder in a curve from a small bump. For example, when the true principal component function has a shoulder, the probability that the empirical principal component function has instead a bump is approximately equal to 1/2. We suggest and describe the performance of bootstrap methods for assessing the strength of extrema. It is shown that the subsample bootstrap is more effective than the standard bootstrap in this regard. A ``bootstrap likelihood is proposed for measuring extremum strength. Exploratory numerical methods are suggested.

Recommendations

• On Properties of Functional Principal Components Analysis
• Principal components analysis of sampled functions
• Kernel-based functional principal components
• Extremal properties of principal curves in the plane
• Principal components for multivariate functional data

Cites work

• scientific article; zbMATH DE number 3124350 (Why is no real title available?)
• scientific article; zbMATH DE number 469124 (Why is no real title available?)
• scientific article; zbMATH DE number 469335 (Why is no real title available?)
• scientific article; zbMATH DE number 2015204 (Why is no real title available?)
• scientific article; zbMATH DE number 795281 (Why is no real title available?)
• scientific article; zbMATH DE number 3211110 (Why is no real title available?)
• A Graphical Technique for Determining the Number of Components in a Mixture of Normals
• A nonlinear PCA based on manifold approximation
• Applied functional data analysis. Methods and case studies
• Asymptotic theory for the principal component analysis of a vector random function: Some applications to statistical inference
• Calibrating the Excess Mass and Dip Tests of Modality
• Curves discrimination: a nonparametric functional approach
• Density estimation under qualitative assumptions in higher dimensions
• Excess Mass Estimates and Tests for Multimodality
• Functional Classification in Hilbert Spaces
• Functional canonical analysis for square integrable stochastic processes
• Functional data analysis.
• Gait analysis and the bootstrap
• Kernel-based functional principal components
• Least squares estimators of the mode of a unimodal regression function
• Linear processes in function spaces. Theory and applications
• Mode testing via the excess mass estimate
• Nonparametric Mixed Effects Models for Unequally Sampled Noisy Curves
• Nonparametric estimation of smoothed principal components analysis of sampled noisy functions
• Nonparametric models for functional data, with application in regression, time series prediction and curve discrimination
• Nonparametric testing of the existence of modes
• On nonparametric kernel estimation of the mode of the regression function in the random design model
• On the asymptotic normality of kernel regression estimators of the mode in the nonparametric random design model.
• Perturbation of spectral subspaces and solution of linear operator equations
• Principal component models for sparse functional data
• Principal components analysis of sampled functions
• Regression Analysis for a Functional Response
• Scale space view of curve estimation.
• SiZer for Exploration of Structures in Curves
• Smoothed functional principal components analysis by choice of norm
• Some asymptotics for multimodality tests based on kernel density estimates
• Statistical analysis of temporal evolution in single-neuron firing rates
• Subsampling
• Test of Significance When Data Are Curves
• The dip test of unimodality
• Varying-coefficient models and basis function approximations for the analysis of repeated measurements
• Weak convergence for the covariance operators of a Hilbertian linear process.
• Étude asymptotique d'un estimateur spline hybride pour le modèle linéaire fonctionnel

Cited in

(1)

• Minimax adaptive tests for the functional linear model

Describes a project that uses

Uses Software

• fda (R)
• SiZer

This page was built for publication: Assessing extrema of empirical principal component functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2500461)