Accéder directement au contenu Accéder directement à la navigation
Article dans une revue

Adaptive estimation of high-dimensional signal-to-noise ratios

Abstract : We consider the equivalent problems of estimating the residual variance, the proportion of explained variance. and the signal strength in a high-dimensional linear regression model with Gaussian random design. Our aim is to understand the impact of not knowing the sparsity of the vector of regression coefficients and not knowing the distribution of the design on minimax estimation rates of.. Depending on the sparsity k of the vector regression coefficients, optimal estimators of. either rely on estimating the vector of regression coefficients or are based on U-type statistics. In the important situation where k is unknown, we build an adaptive procedure whose convergence rate simultaneously achieves the minimax risk over all k up to a logarithmic loss which we prove to be non avoidable. Finally, the knowledge of the design distribution is shown to play a critical role. When the distribution of the design is unknown, consistent estimation of explained variance is indeed possible in much narrower regimes than for known design distribution.
Type de document :
Article dans une revue
Liste complète des métadonnées

Littérature citée [49 références]  Voir  Masquer  Télécharger

https://hal.inrae.fr/hal-02622458
Déposant : Migration Prodinra <>
Soumis le : mardi 26 mai 2020 - 05:43:02
Dernière modification le : jeudi 2 juillet 2020 - 14:03:34

Fichier

publis18-mistea-013_verzelen_a...
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Nicolas Verzelen, Elisabeth Gassiat. Adaptive estimation of high-dimensional signal-to-noise ratios. Bernoulli, Bernoulli Society for Mathematical Statistics and Probability, 2018, 24 (4B), pp.3683-3710. ⟨10.3150/17-BEJ975⟩. ⟨hal-02622458⟩

Partager

Métriques

Consultations de la notice

285

Téléchargements de fichiers

40