Group Bound: Confidence Intervals for Groups of Variables in Sparse High Dimensional Regression Without Assumptions on the Design
From MaRDI portal
Publication:5378142
Abstract: It is in general challenging to provide confidence intervals for individual variables in high-dimensional regression without making strict or unverifiable assumptions on the design matrix. We show here that a "group-bound" confidence interval can be derived without making any assumptions on the design matrix. The lower bound for the regression coefficient of individual variables can be derived via linear programming. The idea also generalises naturally to groups of variables, where we can derive a one-sided confidence interval for the joint effect of a group. While the confidence intervals of individual variables are by the nature of the problem often very wide, it is shown to be possible to detect the contribution of groups of highly correlated predictor variables even when no variable individually shows a significant effect. The assumptions necessary to detect the effect of groups of variables are shown to be weaker than the weakest known assumptions to detect the effect of individual variables.
Recommendations
- The high-dimensional statistical analysis of sparse Group Lasso
- Confidence intervals for sparse penalized regression with random designs
- Fixed-size confidence regions in high-dimensional sparse linear regression models
- Convergence and sparsity of Lasso and group Lasso in high-dimensional generalized linear models
- Confidence intervals for low dimensional parameters in high dimensional linear models
- Confidence intervals for parameters in high-dimensional sparse vector autoregression
- Sparse Group Lasso: Optimal Sample Complexity, Convergence Rate, and Statistical Inference
- PAC-Bayesian risk bounds for group-analysis sparse regression by exponential weighting
- Consistent group selection in high-dimensional linear regression
- Group Lasso estimation of high-dimensional covariance matrices
Cited in
(20)- A bootstrap Lasso+partial ridge method to construct confidence intervals for parameters in high-dimensional sparse linear models
- High-dimensional inference: confidence intervals, \(p\)-values and R-software \texttt{hdi}
- Sparsest representations and approximations of an underdetermined linear system
- Markov Neighborhood Regression for High-Dimensional Inference
- Bootstrapping and sample splitting for high-dimensional, assumption-lean inference
- Debiasing the Lasso: optimal sample size for Gaussian designs
- A sequential rejection testing method for high-dimensional regression with correlated variables
- Prediction regions through inverse regression
- Group inference in high dimensions with applications to hierarchical testing
- Regularized projection score estimation of treatment effects in high-dimensional quantile regression
- The benefit of group sparsity in group inference with de-biased scaled group Lasso
- Goodness-of-Fit Tests for High Dimensional Linear Models
- Two-directional simultaneous inference for high-dimensional models
- A conformal test of linear models via permutation-augmented regressions
- scientific article; zbMATH DE number 7626707 (Why is no real title available?)
- Post-regularization inference for time-varying nonparanormal graphical models
- Inference for high-dimensional varying-coefficient quantile regression
- Spatially relaxed inference on high-dimensional linear models
- High-dimensional simultaneous inference with the bootstrap
- Sparse linear regression models of high dimensional covariates with non-Gaussian outliers and Berkson error-in-variable under heteroscedasticity
This page was built for publication: Group Bound: Confidence Intervals for Groups of Variables in Sparse High Dimensional Regression Without Assumptions on the Design
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5378142)
