Connecting continuum regression with sufficient dimension reduction

Chen and Cook (2010) showed that the classical continuum regression with a nonzero tuning parameter may fail to produce a consistent estimator. In this article we show that the continuum regression with either slightly modified constraints or the tuning parameter being zero can estimate the central subspace consistently. We further provide a general class of continuum regression and connect it with sufficient dimension regression.

MoreLess

Year of publication:	2015
Authors:	Chen, Xin ; Zhu, Li-Ping
Published in:	Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 98.2015, C, p. 44-49
Publisher:	Elsevier
Subject:	Central subspace \| Continuum regression \| Inverse regression \| Sufficient dimension reduction

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10011189322