High dimensional ordinary least squares projection for screening variables
From MaRDI portal
Publication:5378148
DOI10.1111/RSSB.12127zbMATH Open1414.62313arXiv1506.01782OpenAlexW1580111774MaRDI QIDQ5378148FDOQ5378148
Publication date: 12 June 2019
Published in: Journal of the Royal Statistical Society Series B: Statistical Methodology (Search for Journal in Brave)
Abstract: Variable selection is a challenging issue in statistical applications when the number of predictors far exceeds the number of observations . In this ultra-high dimensional setting, the sure independence screening (SIS) procedure was introduced to significantly reduce the dimensionality by preserving the true model with overwhelming probability, before a refined second stage analysis. However, the aforementioned sure screening property strongly relies on the assumption that the important variables in the model have large marginal correlations with the response, which rarely holds in reality. To overcome this, we propose a novel and simple screening technique called the high-dimensional ordinary least-squares projection (HOLP). We show that HOLP possesses the sure screening property and gives consistent variable selection without the strong correlation assumption, and has a low computational complexity. A ridge type HOLP procedure is also discussed. Simulation study shows that HOLP performs competitively compared to many other marginal correlation based methods. An application to a mammalian eye disease data illustrates the attractiveness of HOLP.
Full work available at URL: https://arxiv.org/abs/1506.01782
Recommendations
- Sure Independence Screening for Ultrahigh Dimensional Feature Space
- Ultra-high dimensional variable screening via Gram-Schmidt orthogonalization
- Adaptive model-free sure independence screening
- Dynamic tilted current correlation for high dimensional variable screening
- Ultrahigh dimensional feature screening via projection
lassovariable selectionconsistencyordinary least squareshigh dimensionalitysure independent screeninggeneralized inverseMoore-Penrose inverseforward regressionmarginal correlation
Asymptotic properties of parametric estimators (62F12) Linear regression; mixed models (62J05) Ridge regression; shrinkage estimators (Lasso) (62J07)
Cited In (35)
- Nonparametric augmented probability weighting with sparsity
- Conditional characteristic feature screening for massive imbalanced data
- Covariance-insured screening
- Are Latent Factor Regression and Sparse Regression Adequate?
- Partition-based feature screening for categorical data via RKHS embeddings
- Risk spillover network structure learning for correlated financial assets: a directed acyclic graph approach
- Dynamic tilted current correlation for high dimensional variable screening
- Screen then select: a strategy for correlated predictors in high-dimensional quantile regression
- Network-based feature screening with applications to genome data
- Support recovery of Gaussian graphical model with false discovery rate control
- Cluster feature selection in high-dimensional linear models
- Efficient kernel-based variable selection with sparsistency
- Covariate Information Number for Feature Screening in Ultrahigh-Dimensional Supervised Problems
- A selective overview of feature screening methods with applications to neuroimaging data
- Variable selection for categorical response: a comparative study
- Likelihood Ratio Test in Multivariate Linear Regression: from Low to High Dimension
- Structure learning via unstructured kernel-based M-estimation
- Two‐stage penalized regression screening to detect biomarker–treatment interactions in randomized clinical trials
- Robust group variable screening based on maximum Lq-likelihood estimation
- Learning sparse conditional distribution: an efficient kernel-based approach
- Cross-Trait Prediction Accuracy of Summary Statistics in Genome-Wide Association Studies
- On the dimension effect of regularized linear discriminant analysis
- Factor-Adjusted Regularized Model Selection
- Prior Knowledge Guided Ultra-High Dimensional Variable Screening With Application to Neuroimaging Data
- Scalable inference for high-dimensional precision matrix
- A fast adaptive Lasso for the cox regression via safe screening rules
- Title not available (Why is that?)
- RaSE: A Variable Screening Framework via Random Subspace Ensembles
- Scalable and efficient inference via CPE
- Threshold Selection in Feature Screening for Error Rate Control
- Uniform joint screening for ultra-high dimensional graphical models
- A data-driven approach to conditional screening of high-dimensional variables
- High-dimensional variable screening under multicollinearity
- Interaction screening via canonical correlation
- Grouped variable screening for ultra-high dimensional data for linear model
This page was built for publication: High dimensional ordinary least squares projection for screening variables
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5378148)
