Fourth moments and independent component analysis
From MaRDI portal
Publication:254467
DOI10.1214/15-STS520zbMATH Open1332.62196arXiv1406.4765OpenAlexW3105022862WikidataQ109746490 ScholiaQ109746490MaRDI QIDQ254467FDOQ254467
Authors: Sara Taskinen, Klaus Nordhausen, Hannu Oja, Jari Miettinen 
Publication date: 8 March 2016
Published in: Statistical Science (Search for Journal in Brave)
Abstract: In independent component analysis it is assumed that the components of the observed random vector are linear combinations of latent independent random variables, and the aim is then to find an estimate for a transformation matrix back to these independent components. In the engineering literature, there are several traditional estimation procedures based on the use of fourth moments, such as FOBI (fourth order blind identification), JADE (joint approximate diagonalization of eigenmatrices), and FastICA, but the statistical properties of these estimates are not well known. In this paper various independent component functionals based on the fourth moments are discussed in detail, starting with the corresponding optimization problems, deriving the estimating equations and estimation algorithms, and finding asymptotic statistical properties of the estimates. Comparisons of the asymptotic variances of the estimates in wide independent component models show that in most cases JADE and the symmetric version of FastICA perform better than their competitors.
Full work available at URL: https://arxiv.org/abs/1406.4765
Recommendations
Asymptotic properties of parametric estimators (62F12) Factor analysis and principal components; correspondence analysis (62H25) Estimation in multivariate analysis (62H12)
Cites Work
- Invariant co-ordinate selection (with discussion)
- Approximation Theorems of Mathematical Statistics
- \(R\)-estimation for asymmetric independent component analysis
- Robust Statistics
- Semiparametrically efficient inference based on signed ranks in symmetric independent component models
- Robust m-estimators of multivariate location and scatter
- What is Projection Pursuit?
- Cluster Identification Using Projections
- A Projection Pursuit Algorithm for Exploratory Data Analysis
- Tests of multinormality based on location vectors and scatter matrices
- Eigenvectors of a kurtosis matrix as interesting directions to reveal cluster structure
- Measures of multivariate skewness and kurtosis with applications
- Projection pursuit
- Title not available (Why is that?)
- Title not available (Why is that?)
- Blind separation methods based on Pearson system and its extensions
- Efficient independent component analysis
- Separation of uncorrelated stationary time series using autocovariance matrices
- Consistent noisy independent component analysis
- Title not available (Why is that?)
- Estimation and Testing of Parameters in Multivariate Laplace Distribution
- The Deflation-Based FastICA Estimator: Statistical Analysis Revisited
- 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
- Title not available (Why is that?)
- Characteristics of multivariate distributions and the invariant coordinate system
- Independent component analysis via nonparametric maximum likelihood estimation
- Multivariate skewness and kurtosis measures with an application in ICA
- Robust measures of tail weight
Cited In (31)
- Finite mixtures, projection pursuit and tensor rank: a triangulation
- \(R\)-estimation for asymmetric independent component analysis
- Clustering via finite nonparametric ICA mixture models
- GMM Estimation of Non-Gaussian Structural Vector Autoregression
- Some remarks on Koziol's kurtosis
- Sampling properties of color independent component analysis
- Independent component analysis for tensor-valued data
- Independent component analysis for multivariate functional data
- Fast tensorial JADE
- Large-sample properties of non-stationary source separation for Gaussian signals
- Stationary subspace analysis based on second-order statistics
- Sliced average variance estimation for multivariate time series
- Independent Component Analysis for Compositional Data
- Extracting conditionally heteroskedastic components using independent component analysis
- Title not available (Why is that?)
- New independent component analysis tools for time series
- Identification of structural multivariate GARCH models
- Non-Gaussian component analysis: testing the dimension of the signal subspace
- On the usage of joint diagonalization in multivariate statistics
- Notion of information and independent component analysis.
- Some theoretical properties of two kurtosis matrices, with application to invariant coordinate selection
- Linear Non-Gaussian Component Analysis Via Maximum Likelihood
- On the estimation of entropy in the fastICA algorithm
- On asymptotic properties of the scatter matrix based estimates for complex valued independent component analysis
- Large-sample properties of unsupervised estimation of the linear discriminant using projection pursuit
- Kurtosis removal for data pre-processing
- Numerical Considerations and a new implementation for invariant coordinate selection
- Estimating the number of signals using principal component analysis
- Independent component analysis: a statistical perspective
- Comments on "CuBICA: Independent Component Analysis by Simultaneous third- and Fourth-Order Cumulant Diagonalization
- JADE for Tensor-Valued Observations
Uses Software
This page was built for publication: Fourth moments and independent component analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q254467)
