Optimal upper and lower bounds for the true and empirical excess risks in heteroscedastic least-squares regression
From MaRDI portal
Publication:1950830
DOI10.1214/12-EJS679zbMath1334.62068arXiv1304.6691MaRDI QIDQ1950830
Publication date: 28 May 2013
Published in: Electronic Journal of Statistics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/1304.6691
sup-normheteroscedasticitylower boundsempirical processleast-squares regressionexcess risklocalized basis
Related Items
Slope heuristics and V-Fold model selection in heteroscedastic regression using strongly localized bases ⋮ On concentration for (regularized) empirical risk minimization ⋮ Concentration behavior of the penalized least squares estimator ⋮ Spatial adaptation in heteroscedastic regression: propagation approach ⋮ Optimal model selection in heteroscedastic regression using piecewise polynomial functions ⋮ Set structured global empirical risk minimizers are rate optimal in general dimensions
Cites Work
- Unnamed Item
- Optimal model selection in density estimation
- A high-dimensional Wilks phenomenon
- Risk bounds for statistical learning
- Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
- Gaussian model selection with an unknown variance
- Minimum contrast estimators on sieves: Exponential bounds and rates of convergence
- A Bennett concentration inequality and its application to suprema of empirical processes
- Left concentration inequalities for empirical processes
- Concentration around the mean for maxima of empirical processes
- Optimal aggregation of classifiers in statistical learning.
- Minimal penalties for Gaussian model selection
- Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder)
- Empirical minimization
- Concentration inequalities and asymptotic results for ratio type empirical processes
- On Talagrand's deviation inequalities for product measures
