Convergence and prediction of principal component scores in high-dimensional settings

DOI10.1214/10-AOS821zbMATH Open1204.62097arXiv1211.2970WikidataQ42702618 ScholiaQ42702618MaRDI QIDQ620562FDOQ620562

Seunggeun Lee, Fred A. Wright, Fei Zou

Publication date: 19 January 2011

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: A number of settings arise in which it is of interest to predict Principal Component (PC) scores for new observations using data from an initial sample. In this paper, we demonstrate that naive approaches to PC score prediction can be substantially biased toward 0 in the analysis of large matrices. This phenomenon is largely related to known inconsistency results for sample eigenvalues and eigenvectors as both dimensions of the matrix increase. For the spiked eigenvalue model for random matrices, we expand the generality of these results, and propose bias-adjusted PC score prediction. In addition, we compute the asymptotic correlation coefficient between PC scores from sample and population eigenvectors. Simulation and real data examples from the genetics literature show the improved bias and numerical properties of our estimators.

Full work available at URL: https://arxiv.org/abs/1211.2970

zbMATH Keywords

PCArandom matrixPC scoresPC regression

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Measures of association (correlation, canonical correlation, etc.) (62H20) Eigenvalues, singular values, and eigenvectors (15A18) Random matrices (algebraic aspects) (15B52)

Cites Work

• Principal component analysis.
• Title not available (Why is that?)
• Title not available (Why is that?)
• Prediction by Supervised Principal Components
• On the distribution of the largest eigenvalue in principal components analysis
• Eigenvalues of large sample covariance matrices of spiked population models
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions
• Title not available (Why is that?)
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• PCA consistency in high dimension, low sample size context
• Title not available (Why is that?)
• Title not available (Why is that?)
• Spectrum estimation for large dimensional covariance matrices using random matrix theory
• 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
• Asymptotic Theory for Principal Component Analysis
• Central limit theorems for eigenvalues in a spiked population model
• Statistical eigen-inference from large Wishart matrices
• On the Sampling Theory of Roots of Determinantal Equations
• Principal Components
• Additive Risk Models for Survival Data with High‐Dimensional Covariates

Cited In (19)

• Statistical inference for high-dimension, low-sample-size data
• PCA consistency for the power spiked model in high-dimensional settings
• Modelling functional additive quantile regression using support vector machines approach
• Reconstruction of a low-rank matrix in the presence of Gaussian noise
• Effective PCA for high-dimension, low-sample-size data with noise reduction via geometric representations
• Boundary behavior in high dimension, low sample size asymptotics of PCA
• Isolation-by-distance-and-time in a stepping-stone model
• Convergence of algorithms used for principal component analysis
• Targeted Random Projection for Prediction From High-Dimensional Features
• Computation of ancestry scores with mixed families and unrelated individuals
• Convergence of Sample Eigenvectors of Spiked Population Model
• Asymptotic properties of principal component analysis and shrinkage-bias adjustment under the generalized spiked population model
• Detection, reconstruction, and characterization algorithms from noisy data in multistatic wave imaging
• On the number of principal components in high dimensions
• Adjusting systematic bias in high dimensional principal component scores
• A note on cyclic shift permutation testing for large eigenvalues
• \(e\)PCA: high dimensional exponential family PCA
• Quantile regression estimation of partially linear additive models
• Multidimensional scaling of noisy high dimensional data

Uses Software

• PLINK
• Eigenstrat

   Recommendations

• Convergence of sample eigenvalues, eigenvectors, and principal component scores for ultra-high dimensional data 👍 👎
• A note on the prediction error of principal component regression in high dimensions 👍 👎
• Principal component analysis in very high-dimensional spaces 👍 👎
• On Consistency and Sparsity for Principal Components Analysis in High Dimensions 👍 👎
• Estimating common principal components in high dimensions 👍 👎
• On the number of principal components in high dimensions 👍 👎
• Convergence of algorithms used for principal component analysis 👍 👎
• Geometric consistency of principal component scores for high‐dimensional mixture models and its application 👍 👎
• Principal regression for high dimensional covariance matrices 👍 👎
• Concordance‐based estimation approaches for the optimal sufficient dimension reduction score 👍 👎

This page was built for publication: Convergence and prediction of principal component scores in high-dimensional settings

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q620562)