PCA consistency in high dimension, low sample size context
From MaRDI portal
\(\rho \)-mixingconsistency and strong inconsistencyhigh dimensionlow sample size datanonstandard asymptoticsprincipal component analysissample covariance matrixspiked population model
Eigenvalues, singular values, and eigenvectors (15A18) Asymptotic distribution of eigenvalues, asymptotic theory of eigenfunctions for ordinary differential operators (34L20) Asymptotic properties of parametric estimators (62F12) Factor analysis and principal components; correspondence analysis (62H25)
Abstract: Principal Component Analysis (PCA) is an important tool of dimension reduction especially when the dimension (or the number of variables) is very high. Asymptotic studies where the sample size is fixed, and the dimension grows [i.e., High Dimension, Low Sample Size (HDLSS)] are becoming increasingly relevant. We investigate the asymptotic behavior of the Principal Component (PC) directions. HDLSS asymptotics are used to study consistency, strong inconsistency and subspace consistency. We show that if the first few eigenvalues of a population covariance matrix are large enough compared to the others, then the corresponding estimated PC directions are consistent or converge to the appropriate subspace (subspace consistency) and most other PC directions are strongly inconsistent. Broad sets of sufficient conditions for each of these cases are specified and the main theorem gives a catalogue of possible combinations. In preparation for these results, we show that the geometric representation of HDLSS data holds under general conditions, which includes a -mixing condition and a broad range of sphericity measures of the covariance matrix.
Recommendations
- PCA Consistency for Non-Gaussian Data in High Dimension, Low Sample Size Context
- Consistency of sparse PCA in high dimension, low sample size contexts
- High dimension low sample size asymptotics of robust PCA
- On consistency and sparsity for principal components analysis in high dimensions
- Boundary behavior in high dimension, low sample size asymptotics of PCA
- A general framework for consistency of principal component analysis
- PCA stability and choice of dimensionality
- On asymptotic normality of cross data matrix-based PCA in high dimension low sample size
Cites work
- Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
- Basic properties of strong mixing conditions. A survey and some open questions
- Eigenvalues of large sample covariance matrices of spiked population models
- Geometric Representation of High Dimension, Low Sample Size Data
- scientific article; zbMATH DE number 3347500 (Why is no real title available?)
- Linear Statistical Inference and its Applications
- On Strong Mixing Conditions for Stationary Gaussian Processes
- On the distribution of the largest eigenvalue in principal components analysis
- On Wielandt's inequality and its application to the asymptotic distribution of the eigenvalues of a random symmetric matrix
- Perturbation theory for linear operators.
- Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
- Statistical Significance of Clustering for High-Dimension, Low–Sample Size Data
- The distribution of a statistic used for testing sphericity of normal distributions
- The high-dimension, low-sample-size geometric representation holds under mild conditions
Cited in
(only showing first 100 items - show all)- Integrative sparse principal component analysis
- Sparse sliced inverse regression via Lasso
- On two-sample mean tests under spiked covariances
- Asymptotic performance of PCA for high-dimensional heteroscedastic data
- Continuum directions for supervised dimension reduction
- Distribution-free high-dimensional two-sample tests based on discriminating hyperplanes
- Asymptotics for empirical eigenvalue processes in high-dimensional linear factor models
- Projection pursuit via white noise matrices
- Consistency of sparse PCA in high dimension, low sample size contexts
- Distance-based and RKHS-based dependence metrics in high dimension
- Unbounded largest eigenvalue of large sample covariance matrices: asymptotics, fluctuations and applications
- Subspace rotations for high-dimensional outlier detection
- Clustering by principal component analysis with Gaussian kernel in high-dimension, low-sample-size settings
- Principal components analysis for mixtures with varying concentrations
- A guide for sparse PCA: model comparison and applications
- Double data piling leads to perfect classification
- Consistency of the objective general index in high-dimensional settings
- More about asymptotic properties of some binary classification methods for high dimensional data
- Asymptotic independence of spiked eigenvalues and linear spectral statistics for large sample covariance matrices
- An \({\ell_p}\) theory of PCA and spectral clustering
- Principal component analysis of infinite variance functional data
- Poisson reduced-rank models with sparse loadings
- Some clustering-based exact distribution-free \(k\)-sample tests applicable to high dimension, low sample size data
- Perturbation theory for cross data matrix-based PCA
- Variance variation criterion and consistency in estimating the number of significant signals of high-dimensional PCA
- Sequential testing for structural stability in approximate factor models
- Limiting laws for divergent spiked eigenvalues and largest nonspiked eigenvalue of sample covariance matrices
- Simple and reliable estimators of coefficients of interest in a model with high-dimensional confounding effects
- The limits of the sample spiked eigenvalues for a high-dimensional generalized Fisher matrix and its applications
- Testing for subsphericity when \(n\) and \(p\) are of different asymptotic order
- High dimension low sample size asymptotics of robust PCA
- Distributed estimation of principal eigenspaces
- On asymptotic normality of cross data matrix-based PCA in high dimension low sample size
- On some graph-based two-sample tests for high dimension, low sample size data
- Testing and estimating change-points in the covariance matrix of a high-dimensional time series
- Equality tests of high-dimensional covariance matrices under the strongly spiked eigenvalue model
- Location-invariant tests of homogeneity of large-dimensional covariance matrices
- Inference on high-dimensional mean vectors under the strongly spiked eigenvalue model
- Optimal estimation and rank detection for sparse spiked covariance matrices
- Some high-dimensional one-sample tests based on functions of interpoint distances
- High-dimensional testing for proportional covariance matrices
- An algorithm for deciding the number of clusters and validation using simulated data with application to exploring crop population structure
- Sparse PCA: optimal rates and adaptive estimation
- On some transformations of high dimension, low sample size data for nearest neighbor classification
- ``Virus hunting using radial distance weighted discrimination
- On the optimality of sliced inverse regression in high dimensions
- A high dimensional dissimilarity measure
- On the eigenstructure of covariance matrices with divergent spikes
- Reconstruction of a high-dimensional low-rank matrix
- A general framework for consistency of principal component analysis
- Significance analysis of high-dimensional, low-sample size partially labeled data
- Convergence of sample eigenvalues, eigenvectors, and principal component scores for ultra-high dimensional data
- Overview of object oriented data analysis
- Two-stage procedures for high-dimensional data
- Penalized Orthogonal Iteration for Sparse Estimation of Generalized Eigenvalue Problem
- Sparse principal component analysis and iterative thresholding
- PCA Consistency for Non-Gaussian Data in High Dimension, Low Sample Size Context
- Asymptotic distributions of some test criteria for the mean vector with fewer observations than the dimension
- Correlation tests for high-dimensional data using extended cross-data-matrix methodology
- A nonparametric two-sample test applicable to high dimensional data
- Asymptotics of hierarchical clustering for growing dimension
- PCA and eigen-inference for a spiked covariance model with largest eigenvalues of same asymptotic order
- On the border of extreme and mild spiked models in the HDLSS framework
- Boundary behavior in high dimension, low sample size asymptotics of PCA
- Pattern recognition based on canonical correlations in a high dimension low sample size context
- Statistical inference for high-dimension, low-sample-size data
- A High-Dimensional Two-Sample Test for Non-Gaussian Data under a Strongly Spiked Eigenvalue Model
- A test of sphericity for high-dimensional data and its application for detection of divergently spiked noise
- A survey of high dimension low sample size asymptotics
- Consistent high-dimensional Bayesian variable selection via penalized credible regions
- Two-Step Hypothesis Testing When the Number of Variables Exceeds the Sample Size
- Nonsparse learning with latent variables
- scientific article; zbMATH DE number 7387552 (Why is no real title available?)
- Distance-based outlier detection for high dimension, low sample size data
- Adjusting systematic bias in high dimensional principal component scores
- A classifier under the strongly spiked eigenvalue model in high-dimension, low-sample-size context
- High dimensional asymptotics for the naive Hotelling T2 statistic in pattern recognition
- Joint CLT for top eigenvalues of sample covariance matrices of separable high dimensional long memory processes
- Kernel naive Bayes discrimination for high-dimensional pattern recognition
- Geometric consistency of principal component scores for high-dimensional mixture models and its application
- Binary discrimination methods for high-dimensional data with a geometric representation
- Sparse principal component analysis in Hilbert space
- On consistency and sparsity for principal components analysis in high dimensions
- Principal component analysis (PCA) for high-dimensional data. PCA is dead. Long live PCA
- Using observed confidence levels to perform principal component analyses
- Large covariance estimation by thresholding principal orthogonal complements. With discussion and authors' reply
- Wald Statistics in high-dimensional PCA
- Projected principal component analysis in factor models
- Partial least squares for functional joint models with applications to the Alzheimer's disease neuroimaging initiative study
- A CLT for the LSS of large-dimensional sample covariance matrices with diverging spikes
- Convergence and prediction of principal component scores in high-dimensional settings
- Nonparametric conditional mean testing via an extreme-type statistic in high dimension
- On some high-dimensional two-sample tests based on averages of inter-point distances
- Ridge Regression Under Dense Factor Augmented Models
- Homogeneity tests of covariance for high-dimensional functional data with applications to event segmentation
- On robust estimators of a sphericity measure in high dimension
- Some clustering-based change-point detection methods applicable to high dimension, low sample size data
- A model-based multithreshold method for subgroup identification
- Sparse Principal Component Analysis Based on Least Trimmed Squares
- Double data piling: a high-dimensional solution for asymptotically perfect multi-category classification
This page was built for publication: PCA consistency in high dimension, low sample size context
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1043724)
