Multivariate Functional Principal Component Analysis for Data Observed on Different (Dimensional) Domains

Abstract

Existing approaches for multivariate functional principal component analysis are restricted to data on the same one-dimensional interval. The presented approach focuses on multivariate functional data on different domains that may differ in dimension, e.g. functions and images. The theoretical basis for multivariate functional principal component analysis is given in terms of a Karhunen-Loève Theorem. For the practically relevant case of a finite Karhunen-Loève representation, a relationship between univariate and multivariate functional principal component analysis is established. This offers an estimation strategy to calculate multivariate functional principal components and scores based on their univariate counterparts. For the resulting estimators, asymptotic results are derived. The approach can be extended to finite univariate expansions in general, not necessarily orthonormal bases. It is also applicable for sparse functional data or data with measurement error. A flexible R-implementation is available on CRAN. The new method is shown to be competitive to existing approaches for data observed on a common one-dimensional domain. The motivating application is a neuroimaging study, where the goal is to explore how longitudinal trajectories of a neuropsychological test score covary with FDG-PET brain scans at baseline. Supplementary material, including detailed proofs, additional simulation results and software is available online.

Revised Version. R-Code for the online appendix is available in the .zip file associated with this article in subdirectory "/Software". The software associated with this article is available on CRAN (packages funData and MFPCA)

References in corpus (1)

• Methodology and convergence rates for functional linear regression

Cited by in corpus (42)

• Partial Separability and Functional Graphical Models for Multivariate Gaussian Processes
• Joint Model for Survival and Multivariate Sparse Functional Data with Application to a Study of Alzheimer's Disease
• Intrinsic Riemannian Functional Data Analysis
• On function-on-function regression: Partial least squares approach
• Object-Oriented Software for Functional Data
• Robust Multivariate Functional Control Chart
• Independent component analysis for multivariate functional data
• Clustering multivariate functional data using unsupervised binary trees
• A General Framework for Multivariate Functional Principal Component Analysis of Amplitude and Phase Variation
• Workplace sustainability or financial resilience? Composite-financial resilience index
• A Functional approach for Two Way Dimension Reduction in Time Series
• Sparse Functional Boxplots for Multivariate Curves
• The Impact of Model Assumptions in Scalar-on-Image Regression
• Multivariate functional group sparse regression: functional predictor selection
• Multipopulation mortality modelling and forecasting: The multivariate functional principal component with time weightings approaches
• Change Detection: A functional analysis perspective
• Interpretable Principal Components Analysis for Multilevel Multivariate Functional Data, with Application to EEG Experiments
• Robust Functional Principal Component Analysis for Non-Gaussian Longitudinal Data
• Stein's method, Gaussian processes and Palm measures, with applications to queueing
• Latent Deformation Models for Multivariate Functional Data and Time Warping Separability
• Projection-based two-sample inference for sparsely observed multivariate functional data
• Functional Principal Component Analysis for Extrapolating Multi-stream Longitudinal Data
• Analysing kinematic data from recreational runners using functional data analysis
• Using Persistent Homology Topological Features to Characterize Medical Images: Case Studies on Lung and Brain Cancers
• Deep Time Series Models for Scarce Data
• Fast Covariance Estimation for Multivariate Sparse Functional Data
• Curvature and Torsion estimation of 3D functional data: A geometric approach to build the mean shape under the Frenet Serret framework
• Multivariate Functional Singular Spectrum Analysis Over Different Dimensional Domains
• Filtrated Common Functional Principal Components for Multivariate Functional data
• On the estimation of the number of components in multivariate functional principal component analysis
• FDApy: a Python package for functional data
• Factor-guided estimation of large covariance matrix function with conditional functional sparsity
• Different PCA approaches for vector functional time series with applications to resistive switching processes
• A Tracy-Widom Empirical Estimator For Valid P-values With High-Dimensional Datasets
• Region-Referenced Spectral Power Dynamics of EEG Signals: A Hierarchical Modeling Approach
• A Non-linear Function-on-Function Model for Regression with Time Series Data
• Interpretable discriminant analysis for functional data supported on random nonlinear domains with an application to Alzheimer's disease
• Factor Models for High-Dimensional Functional Time Series
• Functional Data Analysis with Causation in Observational Studies: Covariate Balancing Functional Propensity Score for Functional Treatments
• Benchmarking multi-step methods for the dynamic prediction of survival with numerous longitudinal predictors
• Latent Functional PARAFAC for modeling multidimensional longitudinal data
• Asymmetric canonical correlation analysis of Riemannian and high-dimensional data