Consistent Regression using Data-Dependent Coverings - Archive ouverte HAL Access content directly
Journal Articles Electronic Journal of Statistics Year : 2021

Consistent Regression using Data-Dependent Coverings


In this paper, we introduce a novel method to generate interpretable regression function estimators. The idea is based on called data-dependent coverings. The aim is to extract from the data a covering of the feature space instead of a partition. The estimator predicts the empirical conditional expectation over the cells of the partitions generated from the coverings. Thus, such estimator has the same form as those issued from data-dependent partitioning algorithms. We give sufficient conditions to ensure the consistency, avoiding the sufficient condition of shrinkage of the cells that appears in the former literature. Doing so, we reduce the number of covering elements. We show that such coverings are interpretable and each element of the covering is tagged as significant or insignificant. The proof of the consistency is based on a control of the error of the empirical estimation of conditional expectations which is interesting on its own.
Fichier principal
Vignette du fichier
Consistent Regression using Data-Dependent Coverings.pdf (913.42 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02170687 , version 1 (03-07-2019)
hal-02170687 , version 2 (12-09-2019)
hal-02170687 , version 3 (22-10-2019)
hal-02170687 , version 4 (23-09-2020)
hal-02170687 , version 5 (22-01-2021)



Vincent Margot, Jean-Patrick Baudry, Frédéric Guilloux, Olivier Wintenberger. Consistent Regression using Data-Dependent Coverings. Electronic Journal of Statistics , 2021, 15 (1), ⟨10.1214/21-EJS1806⟩. ⟨hal-02170687v5⟩
118 View
175 Download



Gmail Facebook X LinkedIn More