Uniform convergence rates for nonparametric regression and principal component analysis in functional/longitudinal data

Abstract

We consider nonparametric estimation of the mean and covariance functions for functional/longitudinal data. Strong uniform convergence rates are developed for estimators that are local-linear smoothers. Our results are obtained in a unified framework in which the number of observations within each curve/cluster can be of any rate relative to the sample size. We show that the convergence rates for the procedures depend on both the number of sample curves and the number of observations on each curve. For sparse functional data, these rates are equivalent to the optimal rates in nonparametric regression. For dense functional data, root-n rates of convergence can be achieved with proper choices of bandwidths. We further derive almost sure rates of convergence for principal component analysis using the estimated covariance function. The results are illustrated with simulation studies.

Published in at http://dx.doi.org/10.1214/10-AOS813 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (6)

• Functional linear regression analysis for longitudinal data
• Generalized functional linear models
• Properties of principal component methods for functional and longitudinal data analysis
• Statistical inferences for functional data
• Deciding the dimension of effective dimension reduction space for functional and high-dimensional data
• Nonparametric estimation of correlation functions in longitudinal and spatial data, with application to colon carcinogenesis experiments

Cited by in corpus (49)

• Functional data analysis for density functions by transformation to a Hilbert space
• Modeling and forecasting electricity spot prices: A functional data perspective
• Multivariate varying coefficient model for functional responses
• Review of Functional Data Analysis
• Spatially Varying Coefficient Model for Neuroimaging Data with Jump Discontinuities
• Joint Model for Survival and Multivariate Sparse Functional Data with Application to a Study of Alzheimer's Disease
• Testing for stationarity of functional time series in the frequency domain
• Semiparametric GEE analysis in partially linear single-index models for longitudinal data
• Mean and Covariance Estimation for Functional Snippets
• Multi-dimensional Functional Principal Component Analysis
• Low-Rank Covariance Function Estimation for Multidimensional Functional Data
• Sampled forms of functional PCA in reproducing kernel Hilbert spaces
• Unified Principal Component Analysis for Sparse and Dense Functional Data under Spatial Dependency
• Functional factor analysis for periodic remote sensing data
• Unexpected properties of bandwidth choice when smoothing discrete data for constructing a functional data classifier
• Sparsely Observed Functional Time Series: Estimation and Prediction
• Joint modeling of longitudinal drug using pattern and time to first relapse in cocaine dependence treatment data
• Universal kernel-type estimation of random fields
• Robust optimal estimation of location from discretely sampled functional data
• Highly Irregular Functional Generalized Linear Regression with Electronic Health Records
• Nonparametric Estimation for SDE with Sparsely Sampled Paths: an FDA Perspective
• Simultaneous Confidence Band for Stationary Covariance Function of Dense Functional Data
• Functional Linear Regression with Mixed Predictors
• Functional Lagged Regression with Sparse Noisy Observations
• Simultaneous Confidence Bands for Functional Data Using the Gaussian Kinematic Formula
• Robust Functional Principal Component Analysis for Non-Gaussian Longitudinal Data
• Nonparametric covariance estimation for mixed longitudinal studies, with applications in midlife women's health
• Time-warped growth processes, with applications to the modeling of boom-bust cycles in house prices
• PROFIT: Projection-based Test in Longitudinal Functional Data
• Estimation of the Mean Function of Functional Data via Deep Neural Networks
• Nonparametric Operator-Regularized Covariance Function Estimation for Functional Data
• Partial least squares for sparsely observed curves with measurement errors
• Test of Weak Separability for Spatially Stationary Functional Field
• Kernel estimation of bivariate time-varying coefficient model for longitudinal data with terminal event
• Minimax convergence rate for estimating the Wasserstein barycenter of random measures on the real line
• Sparse Functional Linear Discriminant Analysis
• Adaptive functional principal components analysis
• Clustering of longitudinal curves via a penalized method and EM algorithm
• The Function-on-Scalar LASSO with Applications to Longitudinal GWAS
• Sensible Functional Linear Discriminant Analysis
• Latent function-on-scalar regression models for observed sequences of binary data: a restricted likelihood approach
• Online Estimation for Functional Data
• Homogeneity Tests of Covariance and Change-Points Identification for High-Dimensional Functional Data
• Unified statistical inference for a novel nonlinear dynamic functional/longitudinal data model
• Generalized Gaussian Process Regression Model for Non-Gaussian Functional Data
• Mean and Covariance Estimation for Discretely Observed High-Dimensional Functional Data: Rates of Convergence and Division of Observational Regimes
• Localized Functional Principal Component Analysis
• Fast implementation of partial least squares for function-on-function regression
• Optimal Designs for Longitudinal and Functional Data