Independent component analysis for multivariate functional data

DOI10.1016/J.JMVA.2019.104568zbMATH Open1436.62232arXiv1712.07641OpenAlexW2780870313WikidataQ109772877 ScholiaQ109772877MaRDI QIDQ2293541FDOQ2293541

Authors: Yanyan Li 

Publication date: 5 February 2020

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: We extend two methods of independent component analysis, fourth order blind identification and joint approximate diagonalization of eigen-matrices, to vector-valued functional data. Multivariate functional data occur naturally and frequently in modern applications, and extending independent component analysis to this setting allows us to distill important information from this type of data, going a step further than the functional principal component analysis. To allow the inversion of the covariance operator we make the assumption that the dependency between the component functions lies in a finite-dimensional subspace. In this subspace we define fourth cross-cumulant operators and use them to construct the two novel, Fisher consistent methods for solving the independent component problem for vector-valued functions. Both simulations and an application on a hand gesture data set show the usefulness and advantages of the proposed methods over functional principal component analysis.

Full work available at URL: https://arxiv.org/abs/1712.07641

Recommendations

• Principal components for multivariate functional data
• Multi-dimensional functional principal component analysis
• Multivariate analysis of variance for functional data
• Multivariate functional principal component analysis for data observed on different (dimensional) domains
• Multilevel functional principal component analysis
• Independent Component Analysis for Compositional Data
• Independent Component Analysis and Blind Signal Separation
• Nonparametric independent component analysis
• Variable-Domain Functional Principal Component Analysis

zbMATH Keywords

dimension reductionfunctional principal component analysisHilbert spacecovariance operatorfourth order blind identificationjoint approximate diagonalization of eigenmatrices

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Functional data analysis (62R10)

Cites Work

• Statistical Analysis of Financial Data in S-Plus
• Multivariate Functional Data Visualization and Outlier Detection
• Functional data analysis.
• ggplot2. Elegant graphics for data analysis
• 10.1162/153244303768966085
• Jacobi Angles for Simultaneous Diagonalization
• Linear Non-Gaussian Component Analysis Via Maximum Likelihood
• Multivariate functional principal component analysis: A normalization approach
• Multivariate functional principal component analysis for data observed on different (dimensional) domains
• A Measure of Directional Outlyingness With Applications to Image Data and Video
• Multivariate functional outlier detection
• Model-based clustering for multivariate functional data
• Independent component analysis for tensor-valued data
• Functional factorial \(K\)-means analysis
• Principal components for multivariate functional data
• Crisp and fuzzy k-means clustering algorithms for multivariate functional data
• Dimension-reduced clustering of functional data via subspace separation
• Functional cluster analysis via orthonormalized Gaussian basis expansions and its application
• Clustering of functional data in a low-dimensional subspace
• Fourth moments and independent component analysis
• Consistent noisy independent component analysis
• Deflation-Based FastICA With Adaptive Choices of Nonlinearities
• Performance analysis of the FastICA algorithm and Crame/spl acute/r-rao bounds for linear independent component analysis
• Joint Measures and Cross-Covariance Operators
• Nonlinear sufficient dimension reduction for functional data
• On invariant coordinate system (ICS) functionals
• New independent component analysis tools for time series
• A Nonparametric Graphical Model for Functional Data With Application to Brain Networks Based on fMRI
• On an additive semigraphoid model for statistical networks with application to pathway analysis
• Dimension reduction for functional data based on weak conditional moments
• JADE for Tensor-Valued Observations

Cited In (11)

• Fourth moments and independent component analysis
• Independent component analysis for tensor-valued data
• Independent Component Analysis and Blind Signal Separation
• Characteristic-function-based independent component analysis
• Independent Component Analysis for Compositional Data
• Functional diffusion maps
• On the usage of joint diagonalization in multivariate statistics
• Novel whitening approaches in functional settings
• Robust functional principal component analysis for non-Gaussian longitudinal data
• A global test for heteroscedastic one-way FMANOVA with applications
• Hierarchical independent component analysis: a multi-resolution non-orthogonal data-driven basis

Uses Software

• R
• MASS (R)
• fda (R)
• ggplot2
• JADE
• BSSasymp

This page was built for publication: Independent component analysis for multivariate functional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2293541)