Group regularized estimation under structural hierarchy
From MaRDI portal
Publication:4690971
Abstract: Variable selection for models including interactions between explanatory variables often needs to obey certain hierarchical constraints. The weak or strong structural hierarchy requires that the existence of an interaction term implies at least one or both associated main effects to be present in the model. Lately, this problem has attracted a lot of attention, but existing computational algorithms converge slow even with a moderate number of predictors. Moreover, in contrast to the rich literature on ordinary variable selection, there is a lack of statistical theory to show reasonably low error rates of hierarchical variable selection. This work investigates a new class of estimators that make use of multiple group penalties to capture structural parsimony. We give the minimax lower bounds for strong and weak hierarchical variable selection and show that the proposed estimators enjoy sharp rate oracle inequalities. A general-purpose algorithm is developed with guaranteed convergence and global optimality. Simulations and real data experiments demonstrate the efficiency and efficacy of the proposed approach.
Recommendations
Cites work
- scientific article; zbMATH DE number 5564093 (Why is no real title available?)
- scientific article; zbMATH DE number 4079168 (Why is no real title available?)
- scientific article; zbMATH DE number 47310 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- scientific article; zbMATH DE number 5251637 (Why is no real title available?)
- A lasso for hierarchical interactions
- A unified framework for high-dimensional analysis of \(M\)-estimators with decomposable regularizers
- An iterative algorithm for fitting nonconvex penalized generalized linear models with grouped predictors
- Bayesian variable selection with related predictors
- Consistent group selection in high-dimensional linear regression
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- Interaction screening for ultrahigh-dimensional data
- Model Selection and Estimation in Regression with Grouped Variables
- Oracle inequalities and optimal inference under group sparsity
- Proximal methods for hierarchical sparse coding
- Regularization and Variable Selection Via the Elastic Net
- Simultaneous analysis of Lasso and Dantzig selector
- Sparse spatial autoregressions
- Stationary-sparse causality network learning
- The sparsity and bias of the LASSO selection in high-dimensional linear regression
- Variable selection using adaptive nonlinear interaction structures in high dimensions
- Variable selection with the strong heredity constraint and its oracle property
- Weakly decomposable regularization penalties and structured sparsity
Cited in
(25)- Variable selection with the strong heredity constraint and its oracle property
- High-Dimensional Gaussian Graphical Regression Models with Covariates
- HiQR: an efficient algorithm for high-dimensional quadratic regression with penalties
- A two-stage regularization method for variable selection and forecasting in high-order interaction model
- A Pliable Lasso
- Convex hierarchical testing of interactions
- A General Framework for Identifying Hierarchical Interactions and Its Application to Genomics Data
- Hierarchical sparse modeling: a choice of two group Lasso formulations
- Structured variable selection via prior-induced hierarchical penalty functions
- An analysis of penalized interaction models
- Variable selection for functional linear models with strong heredity constraint
- Efficient kernel-based variable selection with sparsistency
- Structured variable selection and estimation
- scientific article; zbMATH DE number 7071841 (Why is no real title available?)
- Structured learning in time-dependent Cox models
- A First-Order Optimization Algorithm for Statistical Learning with Hierarchical Sparsity Structure
- BOLT-SSI: A Statistical Approach to Screening Interaction Effects for Ultra-High Dimensional Data
- High-dimensional interactions detection with sparse principal Hessian matrix
- scientific article; zbMATH DE number 7306867 (Why is no real title available?)
- A lasso for hierarchical interactions
- Model selection for functional linear regression with hierarchical structure
- Model selection of hierarchically structured covariates using elastic net
- Sequential interaction group selection by the principle of correlation search for high-dimensional interaction models
- The composite absolute penalties family for grouped and hierarchical variable selection
- Coordinate descent based hierarchical interactive Lasso penalized logistic regression and its application to classification problems
This page was built for publication: Group regularized estimation under structural hierarchy
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4690971)
