A NONPARAMETRIC TEST OF THE NONCONVEXITY OF REGRESSION

Authors
DIACK CAT THOMASAGNAN C
Citation
Cat. Diack et C. Thomasagnan, A NONPARAMETRIC TEST OF THE NONCONVEXITY OF REGRESSION, Journal of nonparametric statistics (Print), 9(4), 1998, pp. 335-362
Citations number
13
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
Journal of nonparametric statistics (Print) → ACNP
ISSN journal
10485252
Volume
9
Issue
4
Year of publication
1998
Pages
335 - 362
Database
ISI
SICI code
1048-5252(1998)9:4<335:ANTOTN>2.0.ZU;2-0
Abstract
This paper proposes a nonparametric test of the non-convexity of a smo oth regression function based on least squares or hybrid splines. By a simple formulation of the convexity hypothesis in the lcass of all po lynomial cubic splines, we build a test which has an asymptotic size e qual to the nominal level. It is shown that the test is consistent and is robust to nonnormality. The behavior of the lest under the local a lternatives is studied.