Sparse linear discriminant analysis by thresholding for high dimensional data

DOI10.1214/10-AOS870zbMATH Open1215.62062arXiv1105.3561OpenAlexW3104038823MaRDI QIDQ548558FDOQ548558

Jun Shao, Xinwei Deng, Yazhen Wang, Sijian Wang

Publication date: 29 June 2011

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

Abstract: In many social, economical, biological and medical studies, one objective is to classify a subject into one of several classes based on a set of variables observed from the subject. Because the probability distribution of the variables is usually unknown, the rule of classification is constructed using a training sample. The well-known linear discriminant analysis (LDA) works well for the situation where the number of variables used for classification is much smaller than the training sample size. Because of the advance in technologies, modern statistical studies often face classification problems with the number of variables much larger than the sample size, and the LDA may perform poorly. We explore when and why the LDA has poor performance and propose a sparse LDA that is asymptotically optimal under some sparsity conditions on the unknown parameters. For illustration of application, we discuss an example of classifying human cancer into two classes of leukemia based on a set of 7,129 genes and a training sample of size 72. A simulation is also conducted to check the performance of the proposed method.

Full work available at URL: https://arxiv.org/abs/1105.3561

Recommendations

• Sparse linear discriminant analysis with applications to high dimensional low sample size data
• General sparse multi-class linear discriminant analysis
• A direct estimation approach to sparse linear discriminant analysis
• The Dantzig discriminant analysis with high dimensional data
• A direct approach to sparse discriminant analysis in ultra-high dimensions

zbMATH Keywords

classificationhigh dimensionalitymisclassification ratenormalityoptimal classification rulesparse estimates

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Medical applications (general) (92C50)

Cites Work

• Regularized linear discriminant analysis and its application in microarrays
• Wrappers for feature subset selection
• Title not available (Why is that?)
• Ideal spatial adaptation by wavelet shrinkage
• Title not available (Why is that?)
• Covariance regularization by thresholding
• High-dimensional classification using features annealed independence rules
• Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
• A Selective Overview of Variable Selection in High Dimensional Feature Space (Invited Review Article)
• Title not available (Why is that?)
• On BIC’s selection consistency for discriminant analysis
• Title not available (Why is that?)

Cited In (87)

• A variable selection criterion for linear discriminant rule and its optimality in high dimensional and large sample data
• Variational discriminant analysis with variable selection
• Simultaneous variable selection and class fusion with penalized distance criterion based classifiers
• Hierarchical clustered multiclass discriminant analysis via cross-validation
• Robust Variable and Interaction Selection for Logistic Regression and General Index Models
• A self-calibrated direct approach to precision matrix estimation and linear discriminant analysis in high dimensions
• A procedure of linear discrimination analysis with detected sparsity structure for high-dimensional multi-class classification
• Sparse overlapped linear discriminant analysis
• PCA consistency for the power spiked model in high-dimensional settings
• Generalized alternating direction method of multipliers: new theoretical insights and applications
• Weak signals in high‐dimensional regression: Detection, estimation and prediction
• Generalized quantile treatment effect: a flexible Bayesian approach using quantile ratio smoothing
• High dimensional discrimination analysis via a semiparametric model
• An improved modified cholesky decomposition approach for precision matrix estimation
• Distance-based classifier by data transformation for high-dimension, strongly spiked eigenvalue models
• Sparse semiparametric discriminant analysis
• On two simple and effective procedures for high dimensional classification of general populations
• Penalized Classification using Fisher’s Linear Discriminant
• Network linear discriminant analysis
• Classification with many classes: challenges and pluses
• Title not available (Why is that?)
• New hard-thresholding rules based on data splitting in high-dimensional imbalanced classification
• Ridge-forward quadratic discriminant analysis in high-dimensional situations
• Variable selection of linear programming discriminant estimator
• Classification tree algorithm for grouped variables
• A general framework for tensor screening through smoothing
• Variable Selection in Nonparametric Classification Via Measurement Error Model Selection Likelihoods
• QUADRO: a supervised dimension reduction method via Rayleigh quotient optimization
• A generative approach to modeling data with quantitative and qualitative responses
• Dynamic linear discriminant analysis in high dimensional space
• High dimensional classification for spatially dependent data with application to neuroimaging
• Random matrix theory in statistics: a review
• Optimal feature selection for sparse linear discriminant analysis and its applications in gene expression data
• On false discovery rate thresholding for classification under sparsity
• General sparse multi-class linear discriminant analysis
• A \(U\)-classifier for high-dimensional data under non-normality
• Continuum directions for supervised dimension reduction
• Sparse HDLSS discrimination with constrained data piling
• Optimal classification in sparse Gaussian graphic model
• Efficient computation for differential network analysis with applications to quadratic discriminant analysis
• High dimensional binary classification under label shift: phase transition and regularization
• Regularized linear discriminant analysis and its application in microarrays
• Generalized quadratic discriminant analysis
• On the dimension effect of regularized linear discriminant analysis
• Regularized generalized eigen-decomposition with applications to sparse supervised feature extraction and sparse discriminant analysis
• Title not available (Why is that?)
• Neyman-Pearson classification: parametrics and sample size requirement
• Graph-based sparse linear discriminant analysis for high-dimensional classification
• How can we identify the sparsity structure pattern of high-dimensional data: an elementary statistical analysis to interpretable machine learning
• Optimal functional supervised classification with separation condition
• Geometric classifiers for high-dimensional noisy data
• Innovated interaction screening for high-dimensional nonlinear classification
• Alternating direction method of multipliers for penalized zero-variance discriminant analysis
• Proximal Methods for Sparse Optimal Scoring and Discriminant Analysis
• The sparse group lasso for high-dimensional integrative linear discriminant analysis with application to alzheimer's disease prediction
• Discriminant analysis in small and large dimensions
• Sparse subspace linear discriminant analysis
• Adaptive threshold-based classification of sparse high-dimensional data
• Covariate-Adjusted Tensor Classification in High-Dimensions
• Sparse linear discriminant analysis in structured covariates space
• Varying coefficient linear discriminant analysis for dynamic data
• Sparse Minimum Discrepancy Approach to Sufficient Dimension Reduction with Simultaneous Variable Selection in Ultrahigh Dimension
• Non-parametric shrinkage mean estimation for quadratic loss functions with unknown covariance matrices
• Classification by likelihood accordance functions
• Optimal variable selection in multi-group sparse discriminant analysis
• A review of discriminant analysis in high dimensions
• MBLDA: a novel multiple between-class linear discriminant analysis
• Optimal Linear Discriminant Analysis for High-Dimensional Functional Data
• Title not available (Why is that?)
• Statistical inference under the strongly spiked eigenvalue model
• Least angle regression for model selection
• High-dimensional mean estimation via \(\ell_1\) penalized normal likelihood
• Regularized Linear Programming Discriminant Rule with Folded Concave Penalty for Ultrahigh-Dimensional Data
• Variance ratio screening for ultrahigh dimensional discriminant analysis
• An Efficient Convex Formulation for Reduced-Rank Linear Discriminant Analysis in High Dimensions
• A locally adaptive shrinkage approach to false selection rate control in high-dimensional classification
• Title not available (Why is that?)
• Matrix Linear Discriminant Analysis
• Bayesian decision rules to classification problems
• The Dantzig Discriminant Analysis with High Dimensional Data
• Linear discriminant analysis with sparse and dense signals
• A high-dimensional classification rule using sample covariance matrix equipped with adjusted estimated eigenvalues
• A survey on Neyman-Pearson classification and suggestions for future research
• Classification in high dimension with an Alternative Feature Augmentation via Nonparametrics and Selection (AFANS)
• An Efficient Greedy Search Algorithm for High-Dimensional Linear Discriminant Analysis
• Optimal discriminant analysis in high-dimensional latent factor models
• A portmanteau local feature discrimination approach to the classification with high-dimensional matrix-variate data

Uses Software

• rda

This page was built for publication: Sparse linear discriminant analysis by thresholding for high dimensional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q548558)