On dimension folding of matrix- or array-valued statistical objects

DOI10.1214/09-AOS737zbMATH Open1183.62091arXiv1002.4789OpenAlexW2052211772MaRDI QIDQ2380098FDOQ2380098

Authors: Bing Li, Min Kyung Kim, Naomi S. Altman 

Publication date: 24 March 2010

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

Abstract: We consider dimension reduction for regression or classification in which the predictors are matrix- or array-valued. This type of predictor arises when measurements are obtained for each combination of two or more underlying variables--for example, the voltage measured at different channels and times in electroencephalography data. For these applications, it is desirable to preserve the array structure of the reduced predictor (e.g., time versus channel), but this cannot be achieved within the conventional dimension reduction formulation. In this paper, we introduce a dimension reduction method, to be called dimension folding, for matrix- and array-valued predictors that preserves the array structure. In an application of dimension folding to an electroencephalography data set, we correctly classify 97 out of 122 subjects as alcoholic or nonalcoholic based on their electroencephalography in a cross-validation sample.

Full work available at URL: https://arxiv.org/abs/1002.4789

Recommendations

• Dimension folding PCA and PFC for matrix-valued predictors
• Ensemble sufficient dimension folding methods for analyzing matrix-valued data
• Sufficient dimension folding for a functional of conditional distribution of matrix- or array-valued objects
• Sufficient dimension reduction and prediction in regression
• Dimension reduction in regression without matrix inversion

zbMATH Keywords

sliced inverse regressionelectroencephalographydirectional regressionKronecker envelopesliced average variance estimate

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Multivariate analysis (62H99) Estimation in multivariate analysis (62H12) Graphical methods in statistics (62A09) Biomedical imaging and signal processing (92C55)

Cites Work

• Envelope models for parsimonious and efficient multivariate linear regression
• Sliced Inverse Regression for Dimension Reduction
• Slicing regression: A link-free regression method
• Dimension reduction in regression without matrix inversion
• Dimension reduction for nonelliptically distributed predictors
• Contour regression: a general approach to dimension reduction
• On Directional Regression for Dimension Reduction
• Title not available (Why is that?)
• On the Interpretation of Regression Plots
• On a projective resampling method for dimension reduction with multivariate responses
• Applied multivariate statistical analysis.
• Dimension reduction strategies for analyzing global gene expression data with a response
• Title not available (Why is that?)
• Sufficient Dimension Reduction via Inverse Regression
• Asymptotics for kernel estimate of sliced inverse regression
• Dimension reduction for conditional mean in regression
• On Principal Hessian Directions for Data Visualization and Dimension Reduction: Another Application of Stein's Lemma
• Dimension Reduction for the Conditionalkth Moment in Regression
• Title not available (Why is that?)
• Generalized Inverses, Ridge Regression, Biased Linear Estimation, and Nonlinear Estimation
• The commutation matrix: Some properties and applications
• Using the Bootstrap to Select One of a New Class of Dimension Reduction Methods
• Successive direction extraction for estimating the central subspace in a multiple-index regres\-sion
• Title not available (Why is that?)
• Graphics for Regressions With a Binary Response
• Determining the dimension of iterative Hessian transformation
• Robust dimension reduction based on canonical correlation
• Title not available (Why is that?)
• Estimating the structural dimension of regressions via parametric inverse regression
• Sufficient dimension reduction and graphics in regression
• Augmenting the bootstrap to analyze high dimensional genomic data
• Comments on: Augmenting the bootstrap to analyze high dimensional genomic data

Cited In (47)

• Predictive power of principal components for single-index model and sufficient dimension reduction
• A classifier for multi-dimensional datasets based on Bayesian multiple kernel grouping learning
• A model-free conditional screening approach via sufficient dimension reduction
• Hypothesis Testing of Matrix Graph Model with Application to Brain Connectivity Analysis
• Sliced inverse regression for integrative multi-omics data analysis
• Functional Graphical Models
• Sliced average variance estimation for tensor data
• Tensor sufficient dimension reduction
• Ensemble sufficient dimension folding methods for analyzing matrix-valued data
• Title not available (Why is that?)
• Sufficient dimension folding via tensor inverse regression
• Independent component analysis for tensor-valued data
• Sufficient dimension folding for a functional of conditional distribution of matrix- or array-valued objects
• Hyperparameter estimation for sparse Bayesian learning models
• Latent factor model for multivariate functional data
• Nonparametric Functional Graphical Modeling Through Functional Additive Regression Operator
• Quad folding: a simple idea for the subjective property characterization of large sample sets
• On testing common indices for two multi-index models: a link-free approach
• Covariance estimation via sparse Kronecker structures
• A general framework for tensor screening through smoothing
• A varying coefficient model with matrix valued covariates
• Robust estimator of the correlation matrix with sparse Kronecker structure for a high-dimensional matrix-variate
• On the asymptotic normality and efficiency of Kronecker envelope principal component analysis
• Structured time-dependent inverse regression (STIR)
• Conditional mean dimension reduction for tensor time series
• Low-rank latent matrix-factor prediction modeling for generalized high-dimensional matrix-variate regression
• On sufficient dimension reduction via principal asymmetric least squares
• Tensor sliced inverse regression
• Improved estimation in tensor regression with multiple change-points
• Statistical inference on the significance of rows and columns for matrix-valued data in an additive model
• Dimension Reduction for Fréchet Regression
• Likelihood-Based Dimension Folding on Tensor Data
• A Nonparametric Graphical Model for Functional Data With Application to Brain Networks Based on fMRI
• Trace pursuit variable selection for multi-population data
• Copula Gaussian Graphical Models for Functional Data
• Title not available (Why is that?)
• Fast and Separable Estimation in High-Dimensional Tensor Gaussian Graphical Models
• Tensor Regression with Applications in Neuroimaging Data Analysis
• Model-Free Variable Selection With Matrix-Valued Predictors
• A Penalized Likelihood Method for Classification With Matrix-Valued Predictors
• Higher-order sliced inverse regressions
• Multiple-population shrinkage estimation via sliced inverse regression
• Sparse multivariate functional principal component analysis
• Matrix regression heterogeneity analysis
• Sparse Minimum Discrepancy Approach to Sufficient Dimension Reduction with Simultaneous Variable Selection in Ultrahigh Dimension
• JADE for Tensor-Valued Observations
• A portmanteau local feature discrimination approach to the classification with high-dimensional matrix-variate data

This page was built for publication: On dimension folding of matrix- or array-valued statistical objects

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2380098)