Model selection by resampling penalization

DOI10.1214/08-EJS196zbMath1326.62097arXiv0906.3124OpenAlexW2046044912MaRDI QIDQ1951992

Sylvain Arlot

Publication date: 27 May 2013

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0906.3124

zbMATH Keywords

penalizationmodel selectionresamplingadaptivitynonparametric regressionnonparametric statisticsheteroscedastic dataregressogramexchangeable weighted bootstraphistogram selection

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Nonparametric statistical resampling methods (62G09)

Related Items

Some nonasymptotic results on resampling in high dimension. I: Confidence regions ⋮ Adaptive estimation for an inverse regression model with unknown operator ⋮ Model selection in linear regression using paired bootstrap ⋮ Adaptive partitioning schemes for bipartite ranking ⋮ Model selection: from theory to practice ⋮ Segmentation of the mean of heteroscedastic data via cross-validation ⋮ Spatial adaptation in heteroscedastic regression: propagation approach ⋮ Optimal model selection in heteroscedastic regression using piecewise polynomial functions ⋮ Model selection by resampling penalization ⋮ Optimal model selection in density estimation ⋮ Adaptive density estimation of stationary \(\beta\)-mixing and \(\tau\)-mixing processes ⋮ Estimator selection with respect to Hellinger-type risks ⋮ Margin-adaptive model selection in statistical learning ⋮ Second order correctness of perturbation bootstrap M-estimator of multiple linear regression parameter ⋮ Optimal model selection for density estimation of stationary data under various mixing condi\-tions ⋮ Cytometry inference through adaptive atomic deconvolution ⋮ A survey of cross-validation procedures for model selection ⋮ Estimator selection in the Gaussian setting ⋮ High-dimensional Gaussian model selection on a Gaussian design ⋮ Slope heuristics: overview and implementation ⋮ Theoretical analysis of cross-validation for estimating the risk of the k-Nearest Neighbor classifier ⋮ Estimation and model selection for model-based clustering with the conditional classification likelihood

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Minimax theory of image reconstruction
• Some nonasymptotic results on resampling in high dimension. I: Confidence regions
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Asymptotic optimality for \(C_ p\), \(C_ L\), cross-validation and generalized cross-validation: Discrete index set
• Optimal rates of convergence for nonparametric estimators
• When does bootstrap work! Asymptotic results and simulations
• A rank statistics approach to the consistency of a general bootstrap
• Bootstrap methods: another look at the jackknife
• Risk bounds for model selection via penalization
• Bootstrapping log likelihood and EIC, an extension of AIC
• Weak convergence of dependent empirical measures with application to subsampling in function spaces
• Exchangeably weighted bootstraps of the general empirical process
• Consistency of the generalized bootstrap for degenerate \(U\)-statistics
• The weighted bootstrap
• Approximating the negative moments of the Poisson distribution.
• Information-theoretic determination of minimax rates of convergence
• Smooth discrimination analysis
• Model selection for regression on a fixed design
• Estimation of equifrequency histograms
• Jackknife, bootstrap and other resampling methods in regression analysis
• Complexity regularization via localized random penalties
• On general resampling algorithms and their performance in distribution estimation
• Weak convergence and empirical processes. With applications to statistics
• Simultaneous estimation of the mean and the variance in heteroscedastic Gaussian regression
• Model selection by resampling penalization
• Minimal penalties for Gaussian model selection
• Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder)
• Consistency of cross validation for comparing regression procedures
• Adaptive asymptotically efficient estimation in heteroscedastic nonparametric regression
• Statistical predictor identification
• Concentration of the hypergeometric distribution
• Local Rademacher complexities
• Adapting to Unknown Smoothness via Wavelet Shrinkage
• Bootstrap Model Selection
• Model selection for regression on a random design
• Estimating the Error Rate of a Prediction Rule: Improvement on Cross-Validation
• How Biased is the Apparent Error Rate of a Prediction Rule?
• A comparative study of ordinary cross-validation, v-fold cross-validation and the repeated learning-testing methods
• An optimal selection of regression variables
• The Predictive Sample Reuse Method with Applications
• Bounds on Negative Moments
• Improvements on Cross-Validation: The .632+ Bootstrap Method
• Testing the Fit of a Parametric Function
• Rademacher penalties and structural risk minimization
• Learning Theory
• Learning Theory
• The Relationship between Variable Selection and Data Agumentation and a Method for Prediction
• Some Comments on C P
• Histogram selection in non Gaussian regression
• The bootstrap and Edgeworth expansion
• Combinatorial methods in density estimation
• Gaussian model selection
• Model selection and error estimation