Structured Lasso for regression with matrix covariates

DOI10.5705/SS.2012.033MaRDI QIDQ5413282zbMATH OpenOpenAlexFDO

Authors Chenlei Leng, Junlong Zhao

Publication date 29 April 2014

Published in STATISTICA SINICA (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1108.3755, https://semanticscholar.org/paper/1123f8b7ed79720475a18e350c933dad107c0f80

zbMATH Keywords

high-dimensional datamodel selectionnon-asymptotic boundsrestricted eigenvalues

Mathematics Subject Classification ID

Estimation in multivariate analysis (62H12) Linear regression; mixed models (62J05) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Abstract: The Lasso is one of the most important approaches for parameter estimation and variable selection in high dimensional linear regression. At the heart of its success is the attractive rate of convergence result even when $p$, the dimension of the problem, is much larger than the sample size $n$. In particular, Bickel et al. (2009) showed that this rate, in terms of the $el{l}_1$ norm, is of the order $ssqrt(logp)/n$ for a sparsity index $s$. In this paper, we obtain a new bound on the convergence rate by taking advantage of the distributional information of the model. Under the normality or sub-Gaussian assumption, the rate can be improved to nearly $s/sqrtn$ for certain design matrices. We further outline a general partitioning technique that helps to derive sharper convergence rate for the Lasso. The result is applicable to many covariance matrices suitable for high-dimensional data analysis.

Recommendations

• Sparse covariance thresholding for high-dimensional variable selection
• Regularizing LASSO: a consistent variable selection method
• Variable selection in multivariate linear models with high-dimensional covariance matrix estimation
• Feature selection guided by structural information
• High-dimensional graphs and variable selection with the Lasso

Cited in

(28)

• Autoregressive models for matrix-valued time series
• Covariate-adjusted tensor classification in high dimensions
• Trace regression model with simultaneously low rank and row(column) sparse parameter
• Independent component analysis for tensor-valued data
• Factor models for matrix-valued high-dimensional time series
• Double fused Lasso penalized LAD for matrix regression
• Double fused Lasso regularized regression with both matrix and vector valued predictors
• Sparse matrix linear models for structured high-throughput data
• Lasso ANOVA decompositions for matrix and tensor data
• High-dimensional least square matrix regression
• L2RM: Low-Rank Linear Regression Models for High-Dimensional Matrix Responses
• SLASSO: a scaled LASSO for multicollinear situations
• Model-Free Variable Selection With Matrix-Valued Predictors
• Tensor Canonical Correlation Analysis With Convergence and Statistical Guarantees
• Matrix Autoregressive Spatio-Temporal Models
• Optimal minimax variable selection for large-scale matrix linear regression model
• scientific article; zbMATH DE number 2200484 (Why is no real title available?)
• High-Dimensional Vector Autoregressive Time Series Modeling via Tensor Decomposition
• Statistical inference on the significance of rows and columns for matrix-valued data in an additive model
• A varying coefficient model with matrix valued covariates
• The spatial-temporal lag model of matrix-valued time series and its application
• Projection pursuit Bayesian regression for symmetric matrix predictors
• Low-rank modeling for functional linear regression with two-way functional predictor
• Bayesian scalar-on-network regression with applications to brain functional connectivity
• A Bayesian joint model for mediation analysis with matrix-valued mediators
• Robust estimation of matrix-valued time series autoregressive model
• Robust matrix factor analysis method with adaptive parameter adjustment using Cauchy weighting
• Envelope Matrix Autoregressive Models

This page was built for publication: Structured Lasso for regression with matrix covariates

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5413282)