Robust High-Dimensional Regression with Coefficient Thresholding and Its Application to Imaging Data Analysis
From MaRDI portal
Publication:6154026
DOI10.1080/01621459.2022.2142590arXiv2109.14856OpenAlexW3203908636MaRDI QIDQ6154026FDOQ6154026
Authors:
Publication date: 19 March 2024
Published in: Journal of the American Statistical Association (Search for Journal in Brave)
Abstract: It is of importance to develop statistical techniques to analyze high-dimensional data in the presence of both complex dependence and possible outliers in real-world applications such as imaging data analyses. We propose a new robust high-dimensional regression with coefficient thresholding, in which an efficient nonconvex estimation procedure is proposed through a thresholding function and the robust Huber loss. The proposed regularization method accounts for complex dependence structures in predictors and is robust against outliers in outcomes. Theoretically, we analyze rigorously the landscape of the population and empirical risk functions for the proposed method. The fine landscape enables us to establish both {statistical consistency and computational convergence} under the high-dimensional setting. The finite-sample properties of the proposed method are examined by extensive simulation studies. An illustration of real-world application concerns a scalar-on-image regression analysis for an association of psychiatric disorder measured by the general factor of psychopathology with features extracted from the task functional magnetic resonance imaging data in the Adolescent Brain Cognitive Development study.
Full work available at URL: https://arxiv.org/abs/2109.14856
Cites Work
- Nearly unbiased variable selection under minimax concave penalty
- Phase retrieval via Wirtinger flow: theory and algorithms
- The Adaptive Lasso and Its Oracle Properties
- Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
- Title not available (Why is that?)
- One-step sparse estimates in nonconcave penalized likelihood models
- Title not available (Why is that?)
- Sharp Thresholds for High-Dimensional and Noisy Sparsity Recovery Using $\ell _{1}$-Constrained Quadratic Programming (Lasso)
- Model Selection and Estimation in Regression with Grouped Variables
- Robust Estimation of a Location Parameter
- A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers
- Gradient methods for minimizing composite functions
- Asymptotic Equivalence of Regularization Methods in Thresholded Parameter Space
- Risk bounds for model selection via penalization
- Statistical consistency and asymptotic normality for high-dimensional robust \(M\)-estimators
- Restricted strong convexity and weighted matrix completion: optimal bounds with noise
- A general theory of concave regularization for high-dimensional sparse estimation problems
- The statistical analysis of fMRI data
- Mini-batch stochastic approximation methods for nonconvex stochastic composite optimization
- Strong oracle optimality of folded concave penalized estimation
- Spatial Bayesian variable selection and grouping for high-dimensional scalar-on-image regression
- On robust regression with high-dimensional predictors
- Optimal rates of convergence for noisy sparse phase retrieval via thresholded Wirtinger flow
- The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions
- Support recovery without incoherence: a case for nonconvex regularization
- The landscape of empirical risk for nonconvex losses
- I-LAMM for sparse learning: simultaneous control of algorithmic complexity and statistical error
- Scalar-on-image regression via the soft-thresholded Gaussian process
- Hard thresholding regression
This page was built for publication: Robust High-Dimensional Regression with Coefficient Thresholding and Its Application to Imaging Data Analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6154026)
