Automatic selection by penalized asymmetric L q -norm in a high-dimensional model with grouped variables
From MaRDI portal
Publication:6083206
Abstract: The paper focuses on the automatic selection of the grouped explanatory variables in an high-dimensional model, when the model errors are asymmetric. After introducing the model and notations, we define the adaptive group LASSO expectile estimator for which we prove the oracle properties: the sparsity and the asymptotic normality. Afterwards, the results are generalized by considering the asymmetric -norm loss function. The theoretical results are obtained in several cases with respect to the number of variable groups. This number can be fixed or dependent on the sample size , with the possibility that it is of the same order as . Note that these new estimators allow us to consider weaker assumptions on the data and on the model errors than the usual ones. Simulation study demonstrates the competitive performance of the proposed penalized expectile regression, especially when the samples size is close to the number of explanatory variables and model errors are asymmetrical. An application on air pollution data is considered.
Recommendations
- Consistent tuning parameter selection in high-dimensional group-penalized regression
- Group selection in high-dimensional partially linear additive models
- Group variable selection via \(\ell_{p,0}\) regularization and application to optimal scoring
- Consistent group selection in high-dimensional linear regression
- Model selection in high-dimensional quantile regression with seamless L₀ penalty
- Variable Selection in Penalized Model‐Based Clustering Via Regularization on Grouped Parameters
- Adaptive group Lasso selection in quantile models
- A convex-Nonconvex strategy for grouped variable selection
- Variable selection in high-dimensional linear model with possibly asymmetric errors
- Group variable selection via a hierarchical lasso and its oracle property
Cites work
- A note on adaptive group Lasso
- Adaptive elastic-net selection in a quantile model with diverging number of variable groups
- Adaptive group Lasso for high-dimensional generalized linear models
- Adaptive group Lasso selection in quantile models
- Adaptive Lasso estimators for ultrahigh dimensional generalized linear models
- Adaptive sparse group LASSO in quantile regression
- Asymptotics for Lasso-type estimators.
- Conditional \(L_ p\)-quantiles and their application to the testing of symmetry in non-parametric regression
- Expectile and quantile regression—David and Goliath?
- Expectile asymptotics
- Expectile regression for analyzing heteroscedasticity in high dimension
- Extreme M-quantiles as risk measures: from \(L^{1}\) to \(L^{p}\) optimization
- On the asymptotics of constrained \(M\)-estimation
- On the oracle property of adaptive group Lasso in high-dimensional linear models
- Overlapping group lasso for high-dimensional generalized linear models
- Penalized expectile regression: an alternative to penalized quantile regression
- Penalized high-dimensional M-quantile regression: from L^1 to L^p optimization
- The Adaptive Lasso and Its Oracle Properties
- Variable selection in high-dimensional linear model with possibly asymmetric errors
- Variable selection in quantile regression
This page was built for publication: Automatic selection by penalized asymmetric L q -norm in a high-dimensional model with grouped variables
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6083206)
