Efficient estimation of conditional covariance matrices for dimension reduction

Abstract : Let X ∈ Rp andY ∈ R. In this paper,we propose an estimator of the conditional covariancematrix, Cov(E[X|Y]), in an inverse regression setting. Based on the estimation of a quadratic functional, this methodology provides an efficient estimator from a semi parametric point of view.We consider a functional Taylor expansion of Cov(E[X|Y]) under some mild conditions and the effect of using an estimate of the unknown joint distribution. The asymptotic properties of this estimator are also provided.
Document type :
Journal articles
Complete list of metadatas

https://hal-ifp.archives-ouvertes.fr/hal-01564965
Contributor : Françoise Bertrand <>
Submitted on : Wednesday, July 19, 2017 - 12:07:30 PM
Last modification on : Thursday, October 17, 2019 - 8:53:24 AM

Identifiers

Citation

Sébastien da Veiga, Jean-Michel Loubes, Maikol Solís. Efficient estimation of conditional covariance matrices for dimension reduction. Communications in Statistics - Theory and Methods, Taylor & Francis, 2017, 46 (9), pp.4403 - 4424. ⟨10.1080/03610926.2015.1083109⟩. ⟨hal-01564965⟩

Share

Metrics

Record views

188