Open Access
Translator Disclaimer
2022 The ensemble conditional variance estimator for sufficient dimension reduction
Lukas Fertl, Efstathia Bura
Author Affiliations +
Electron. J. Statist. 16(1): 1595-1634 (2022). DOI: 10.1214/22-EJS1994

Abstract

Ensemble Conditional Variance Estimation (ECVE) is a novel sufficient dimension reduction (SDR) method in regressions with continuous response and predictors. ECVE applies to general non-additive error regression models and operates under the assumption that the predictors can be replaced by a lower dimensional projection without loss of information. It is a semiparametric forward regression model-based exhaustive sufficient dimension reduction estimation method that is shown to be consistent under mild assumptions. ECVE outperforms central subspace mean average variance estimation (csMAVE), its main competitor, under several simulation settings and in a benchmark data set analysis.

Funding Statement

L. Fertl was supported by Austrian Science Fund (FWF P 30690-N35). E. Bura was supported by Austrian Science Fund (FWF P 30690-N35) and Vienna Science and Technology Fund (WWTF ICT19-018)

Acknowledgments

The authors thank Daniel Kapla for his programming assistance. Daniel Kapla also co-authored the CVE R package that implements the proposed method.

Citation

Download Citation

Lukas Fertl. Efstathia Bura. "The ensemble conditional variance estimator for sufficient dimension reduction." Electron. J. Statist. 16 (1) 1595 - 1634, 2022. https://doi.org/10.1214/22-EJS1994

Information

Received: 1 March 2021; Published: 2022
First available in Project Euclid: 7 March 2022

Digital Object Identifier: 10.1214/22-EJS1994

Keywords: central subspace , ensembles , linear sufficient reduction , regression , semi-parametric

JOURNAL ARTICLE
40 PAGES


SHARE
Vol.16 • No. 1 • 2022
Back to Top