Dynamic linear discriminant analysis in high dimensional space
From MaRDI portal
Publication:2295032
Abstract: High-dimensional data that evolve dynamically feature predominantly in the modern data era. As a partial response to this, recent years have seen increasing emphasis to address the dimensionality challenge. However, the non-static nature of these datasets is largely ignored. This paper addresses both challenges by proposing a novel yet simple dynamic linear programming discriminant (DLPD) rule for binary classification. Different from the usual static linear discriminant analysis, the new method is able to capture the changing distributions of the underlying populations by modeling their means and covariances as smooth functions of covariates of interest. Under an approximate sparse condition, we show that the conditional misclassification rate of the DLPD rule converges to the Bayes risk in probability uniformly over the range of the variables used for modeling the dynamics, when the dimensionality is allowed to grow exponentially with the sample size. The minimax lower bound of the estimation of the Bayes risk is also established, implying that the misclassification rate of our proposed rule is minimax-rate optimal. The promising performance of the DLPD rule is illustrated via extensive simulation studies and the analysis of a breast cancer dataset.
Recommendations
- High Dimensional Linear Discriminant Analysis: Optimality, Adaptive Algorithm and Missing Data
- High dimensional discrimination analysis via a semiparametric model
- High-dimensional linear discriminant analysis using nonparametric methods
- A direct estimation approach to sparse linear discriminant analysis
- Fast rate of convergence in high-dimensional linear discriminant analysis
Cites work
- A direct approach for sparse quadratic discriminant analysis
- A direct approach to sparse discriminant analysis in ultra-high dimensions
- A direct estimation approach to sparse linear discriminant analysis
- A new reduced-rank linear discriminant analysis method and its applications
- A road to classification in high dimensional space: the regularized optimal affine discriminant
- Bernstein inequality and moderate deviations under strong mixing conditions
- Choice of neighbor order in nearest-neighbor classification
- Choosing Between Logistic Regression and Discriminant Analysis
- Confidence intervals for high-dimensional linear regression: minimax rates and adaptivity
- Convergence of estimates under dimensionality restrictions
- Discriminant analysis through a semiparametric model
- High dimensional discrimination analysis via a semiparametric model
- scientific article; zbMATH DE number 991833 (Why is no real title available?)
- scientific article; zbMATH DE number 1964693 (Why is no real title available?)
- Innovated interaction screening for high-dimensional nonlinear classification
- Introduction to nonparametric estimation
- Joint estimation of multiple graphical models
- Multiclass sparse discriminant analysis
- Multivariate local polynomial kernel estimators: leading bias and asymptotic distribution
- Optimal classification in sparse Gaussian graphic model
- Optimal rates of convergence for covariance matrix estimation
- Penalized classification using Fisher's linear discriminant
- Probability inequalities.
- QUADRO: a supervised dimension reduction method via Rayleigh quotient optimization
- Some theory for Fisher's linear discriminant function, `naive Bayes', and some alternatives when there are many more variables than observations
- Sparse linear discriminant analysis by thresholding for high dimensional data
- Sparsifying the Fisher linear discriminant by rotation
- The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
- The Efficiency of Logistic Regression Compared to Normal Discriminant Analysis
- The location model for mixtures of categorical and continuous variables
- Two sample tests for high-dimensional covariance matrices
- Uniform in bandwidth consistency of kernel-type function estimators
Cited in
(9)- Varying coefficient linear discriminant analysis for dynamic data
- Tuning selection for two-scale kernel density estimators
- High Dimensional Linear Discriminant Analysis: Optimality, Adaptive Algorithm and Missing Data
- Variable-dependent partial dimension reduction
- Optimal discriminant analysis in high-dimensional latent factor models
- Multiclass sparse discriminant analysis incorporating graphical structure among predictors
- Dynamic Supervised Principal Component Analysis for Classification
- Kernelized Discriminant Analysis for Joint Modeling of Multivariate Categorical Responses
- A self-calibrated direct approach to precision matrix estimation and linear discriminant analysis in high dimensions
This page was built for publication: Dynamic linear discriminant analysis in high dimensional space
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2295032)
