Model selection via testing: an alternative to (penalized) maximum likelihood estimators.

DOI10.1016/j.anihpb.2005.04.004zbMath1333.62094MaRDI QIDQ2490800

Lucien Birgé

Publication date: 18 May 2006

Published in: Annales de l'Institut Henri Poincaré. Probabilités et Statistiques (Search for Journal in Brave)

Full work available at URL: http://www.numdam.org/item?id=AIHPB_2006__42_3_273_0

zbMATH Keywords

robustness; model selection; maximum likelihood; minimax risk; metric dimension; robust tests; aggregation of estimators

Mathematics Subject Classification ID

62G05: Nonparametric estimation

62F35: Robustness and adaptive procedures (parametric inference)

Related Items

Estimating the intensity of a random measure by histogram type estimators, Linear and convex aggregation of density estimators, Parameter tuning in pointwise adaptation using a propagation approach, From local kernel to nonlocal multiple-model image denoising, Convergence rates of posterior distributions for non iid observations, A universal procedure for aggregating estimators, Model selection for Gaussian regression with random design, Statistical estimation with model selection, Aggregation for Gaussian regression, Simultaneous adaptation to the margin and to complexity in classification, Optimal rates of aggregation in classification under low noise assumption