Effective PCA for high-dimension, low-sample-size data with noise reduction via geometric representations

DOI10.1016/J.JMVA.2011.09.002zbMATH Open1236.62065arXiv1503.04525OpenAlexW2060519623MaRDI QIDQ764487FDOQ764487

Makoto Aoshima, Kazuyoshi Yata

Publication date: 13 March 2012

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

Abstract: In this paper, we consider clustering based on principal component analysis (PCA) for high-dimension, low-sample-size (HDLSS) data. We give theoretical reasons why PCA is effective for clustering HDLSS data. First, we derive a geometric representation of HDLSS data taken from a two-class mixture model. With the help of the geometric representation, we give geometric consistency properties of sample principal component scores in the HDLSS context. We develop ideas of the geometric representation and geometric consistency properties to multiclass mixture models. We show that PCA can classify HDLSS data under certain conditions in a surprisingly explicit way. Finally, we demonstrate the performance of the clustering by using microarray data sets.

Full work available at URL: https://arxiv.org/abs/1503.04525

Recommendations

• Effective PCA for high-dimension, low-sample-size data with singular value decomposition of cross data matrix
• Statistical inference for high-dimension, low-sample-size data
• PCA consistency in high dimension, low sample size context
• Effective methodologies for high-dimensional data
• Boundary behavior in high dimension, low sample size asymptotics of PCA

zbMATH Keywords

consistencydiscriminant analysisprincipal components analysiseigenvalue distributionHDLSSinverse matrixnoise reduction

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Asymptotic distribution theory in statistics (62E20) Estimation in multivariate analysis (62H12) Eigenvalues, singular values, and eigenvectors (15A18) Set-valued maps in general topology (54C60)

Cites Work

• On the distribution of the largest eigenvalue in principal components analysis
• Eigenvalues of large sample covariance matrices of spiked population models
• Title not available (Why is that?)
• PCA consistency in high dimension, low sample size context
• Basic properties of strong mixing conditions. A survey and some open questions
• Multivariate Theory for Analyzing High Dimensional Data
• Convergence and prediction of principal component scores in high-dimensional settings
• Geometric Representation of High Dimension, Low Sample Size Data
• Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
• The high-dimension, low-sample-size geometric representation holds under mild conditions
• On Strong Mixing Conditions for Stationary Gaussian Processes
• PCA Consistency for Non-Gaussian Data in High Dimension, Low Sample Size Context
• Effective PCA for high-dimension, low-sample-size data with singular value decomposition of cross data matrix
• Intrinsic dimensionality estimation of high-dimension, low sample size data with \(D\)-asymptotics
• Comparison of Discrimination Methods for High Dimensional Data
• Minimum distance classification rules for high dimensional data

Cited In (47)

• Estimation of linear functional of large spectral density matrix and application to Whittle's approach
• A test of sphericity for high-dimensional data and its application for detection of divergently spiked noise
• Statistical inference for high-dimension, low-sample-size data
• Title not available (Why is that?)
• Two-Stage Procedures for High-Dimensional Data
• Limiting laws for divergent spiked eigenvalues and largest nonspiked eigenvalue of sample covariance matrices
• A distance-based, misclassification rate adjusted classifier for multiclass, high-dimensional data
• PCA consistency for the power spiked model in high-dimensional settings
• A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model
• Reconstruction of a high-dimensional low-rank matrix
• A survey of high dimension low sample size asymptotics
• Reconstruction of a low-rank matrix in the presence of Gaussian noise
• Distance-based classifier by data transformation for high-dimension, strongly spiked eigenvalue models
• Asymptotic properties of the first principal component and equality tests of covariance matrices in high-dimension, low-sample-size context
• Polynomial whitening for high-dimensional data
• Using visual statistical inference to better understand random class separations in high dimension, low sample size data
• Location-invariant tests of homogeneity of large-dimensional covariance matrices
• Statistical inference under the strongly spiked eigenvalue model
• Asymptotics of hierarchical clustering for growing dimension
• Binary discrimination methods for high-dimensional data with a geometric representation
• Overview of object oriented data analysis
• Correlation tests for high-dimensional data using extended cross-data-matrix methodology
• Discussion on “Two-Stage Procedures for High-Dimensional Data” by Makoto Aoshima and Kazuyoshi Yata
• Robust PCA for high‐dimensional data based on characteristic transformation
• Perturbation theory for cross data matrix-based PCA
• Equality tests of high-dimensional covariance matrices under the strongly spiked eigenvalue model
• Effective PCA for high-dimension, low-sample-size data with singular value decomposition of cross data matrix
• Inference on high-dimensional mean vectors with fewer observations than the dimension
• Authors' Response
• Analysis of high-dimensional one group repeated measures designs
• Clustering by principal component analysis with Gaussian kernel in high-dimension, low-sample-size settings
• Geometric classifiers for high-dimensional noisy data
• CORRELATION MATRIX OF EQUI-CORRELATED NORMAL POPULATION: FLUCTUATION OF THE LARGEST EIGENVALUE, SCALING OF THE BULK EIGENVALUES, AND STOCK MARKET
• On asymptotic normality of cross data matrix-based PCA in high dimension low sample size
• Inference on high-dimensional mean vectors under the strongly spiked eigenvalue model
• Hypothesis tests for high-dimensional covariance structures
• Double data piling leads to perfect classification
• Intrinsic dimensionality estimation of high-dimension, low sample size data with \(D\)-asymptotics
• Semiparametric estimation of the high-dimensional elliptical distribution
• Equality tests of covariance matrices under a low-dimensional factor structure
• Test for high-dimensional outliers with principal component analysis
• Consistency of the objective general index in high-dimensional settings
• More about asymptotic properties of some binary classification methods for high dimensional data
• High-dimensional hypothesis testing for allometric extension model
• A classifier under the strongly spiked eigenvalue model in high-dimension, low-sample-size context
• Asymptotic independence of spiked eigenvalues and linear spectral statistics for large sample covariance matrices
• Title not available (Why is that?)

This page was built for publication: Effective PCA for high-dimension, low-sample-size data with noise reduction via geometric representations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q764487)