Discriminant analysis in small and large dimensions

DOI10.1090/TPMS/1096zbMATH Open1446.62136arXiv1705.02826OpenAlexW2976974208MaRDI QIDQ5117960FDOQ5117960

Nestor Parolya, Author name not available (Why is that?), Taras Bodnar, Stepan Mazur

Publication date: 26 August 2020

Published in: Theory of Probability and Mathematical Statistics (Search for Journal in Brave)

Abstract: We study the distributional properties of the linear discriminant function under the assumption of normality by comparing two groups with the same covariance matrix but different mean vectors. A stochastic representation for the discriminant function coefficients is derived which is then used to obtain their asymptotic distribution under the high-dimensional asymptotic regime. We investigate the performance of the classification analysis based on the discriminant function in both small and large dimensions. A stochastic representation is established which allows to compute the error rate in an efficient way. We further compare the calculated error rate with the optimal one obtained under the assumption that the covariance matrix and the two mean vectors are known. Finally, we present an analytical expression of the error rate calculated in the high-dimensional asymptotic regime. The finite-sample properties of the derived theoretical results are assessed via an extensive Monte Carlo study.

Full work available at URL: https://arxiv.org/abs/1705.02826

Recommendations

• Discriminant analysis of high-dimensional data over limited samples
• Asymptotic results in canonical discriminant analysis when the dimension is large compared to the sample size
• An optimal discriminant rule in the class of linear and quadratic discriminant functions for large dimension and samples
• Asymptotic probabilities of misclassification of two discriminant functions in cases of high dimensional data
• Effect of dimensionality on discrimination

zbMATH Keywords

discriminant functionlarge-dimensional asymptoticsrandom matrix theorystochastic representationclassification analysis

Mathematics Subject Classification ID

Exact distribution theory in statistics (62E15) Multivariate distribution of statistics (62H10) Asymptotic distribution theory in statistics (62E20) Random matrices (probabilistic aspects) (60B20) Central limit and other weak theorems (60F05)

Cites Work

• Title not available (Why is that?)
• High dimensional covariance matrix estimation using a factor model
• Title not available (Why is that?)
• Large Covariance Estimation by Thresholding Principal Orthogonal Complements
• Spectral analysis of large dimensional random matrices
• Bayesian estimation of the global minimum variance portfolio
• Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
• On the distribution of the largest eigenvalue in principal components analysis
• A Direct Estimation Approach to Sparse Linear Discriminant Analysis
• On the strong convergence of the optimal linear shrinkage estimator for large dimensional covariance matrix
• Sparse linear discriminant analysis by thresholding for high dimensional data
• A test for the weights of the global minimum variance portfolio in an elliptical model
• Applied multivariate statistical analysis.
• A Constrainedℓ₁Minimization Approach to Sparse Precision Matrix Estimation
• Title not available (Why is that?)
• Asymptotic theory of statistics and probability
• Methods of multivariate analysis
• Adaptive Thresholding for Sparse Covariance Matrix Estimation
• Tests for High-Dimensional Covariance Matrices
• Limiting laws of coherence of random matrices with applications to testing covariance structure and construction of compressed sensing matrices
• Corrections to LRT on large-dimensional covariance matrix by RMT
• Computational Statistics
• Noisy matrix decomposition via convex relaxation: optimal rates in high dimensions
• An exact test about the covariance matrix
• Direct shrinkage estimation of large dimensional precision matrix
• Elliptically Contoured Models in Statistics and Portfolio Theory
• Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions
• Bayesian estimation of the efficient frontier
• Asymptotic aproximations for EPMC’s of the linear and the quadratic discriminant functions when the sample sizes and the dimension are large
• On the product of inverse Wishart and normal distributions with applications to discriminant analysis and portfolio theory
• Properties of the singular, inverse and generalized inverse partitioned Wishart distributions
• Title not available (Why is that?)
• Title not available (Why is that?)
• Comparison of Discrimination Methods for High Dimensional Data
• Exact and asymptotic tests on a factor model in low and large dimensions with applications
• Testing for independence of large dimensional vectors
• BAYESIAN INFERENCE FOR THE TANGENT PORTFOLIO
• Central limit theorems for functionals of large sample covariance matrix and mean vector in matrix‐variate location mixture of normal distributions

Cited In (14)

• On the mean and variance of the estimated tangency portfolio weights for small samples
• Distribution of the product of a Wishart matrix and a normal vector
• Title not available (Why is that?)
• Bayesian estimation for misclassification rate in linear discriminant analysis
• Discriminant analysis with a stochastic supervisor
• A study on three linear discriminant analysis based methods in small sample size problem
• Kick-one-out-based variable selection method for Euclidean distance-based classifier in high-dimensional settings
• Title not available (Why is that?)
• High-dimensional asymptotic results for EPMCs of W- and Z- rules
• Asymptotic results in canonical discriminant analysis when the dimension is large compared to the sample size
• Asymptotic probabilities of misclassification of two discriminant functions in cases of high dimensional data
• Higher order moments of the estimated tangency portfolio weights
• Title not available (Why is that?)
• Effect of dimensionality on discrimination

This page was built for publication: Discriminant analysis in small and large dimensions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5117960)