Abstract:We propose a nonparametric method to perform functional principal components analysis for the case of sparse longitudinal data. The method aims at irregularly spaced longitudinal data, where the number of repeated measurements available per subject is small. In contrast, classical functional data analysis requires a large number of regularly spaced measurements per subject. We assume that the repeated measurements are located randomly with a random number of repetitions for each subject and are determined by an underlying smooth random (subject-specific) trajectory plus measurement errors. Basic elements of our approach are the parsimonious estimation of the covariance structure and mean function of the trajectories, and the estimation of the variance of the measurement errors. The eigenfunction basis is estimated from the data, and functional principal components score estimates are obtained by a conditioning step. This conditional estimation method is conceptually simple and straightforward to implement. A key step is the derivation of asymptotic consistency and distribution results under mild conditions, using tools from functional analysis. Functional data analysis for sparse longitudinal data enables prediction of individual smooth trajectories even if only one or few measurements are available for a subject. Asymptotic pointwise and simultaneous confidence bands are obtained for predicted individual trajectories, based on asymptotic distributions, for simultaneous bands under the assumption of a finite number of components. Model selection techniques, such as the Akaike information criterion, are used to choose the model dimension corresponding to the number of eigenfunctions in the model. The methods are illustrated with a simulation study, longitudinal CD4 data for a sample of AIDS patients, and time-course gene expression data for the yeast cell cycle.

STATISTICAL INFERENCE FOR FUNCTIONAL TIME SERIES

Functional limit theorem for moving average processes generated by dependent random variables

Statistical Inference for Functional Time Series: Autocovariance Function

Nonparametric M-estimation for Functional Stationary Ergodic Data

Simultaneous Inference for the Mean Function Based on Dense Functional Data

Statistica Sinica Preprint No : SS-2021-0121 Title Statistical Inference for Functional Time Series : Autocovariance Function

A Least Squares Estimator for Lévy-driven Moving Averages Based on Discrete Time Observations

Statistical inference in partially time-varying coefficient models

Inference for Dependent Error Functional Data with Application to Event-Related Potentials

Simultaneous Confidence Band for Stationary Covariance Function of Dense Functional Data

Statistical inferences for functional data

Statistical inference in functional semiparametric spatial autoregressive model

Adaptive estimation of irregular mean and covariance functions

Oracally Efficient Estimation for Dense Functional Data with Holiday Effects

Online Statistical Inference for Nonlinear Stochastic Approximation with Markovian Data

Oracle-efficient estimation for functional data error distribution with simultaneous confidence band

FUSED KERNEL-SPLINE SMOOTHING FOR REPEATEDLY MEASURED OUTCOMES IN A GENERALIZED PARTIALLY LINEAR MODEL WITH FUNCTIONAL SINGLE INDEX

Functional Data Analysis for Sparse Longitudinal Data

Asymptotic Distributions of Nonparametric Regression Estimators for Longitudinal or Functional Data

Online Estimation for Functional Data

Simultaneous inference for monotone and smoothly time-varying functions under complex temporal dynamics