Goodness of fit test for isotonic regression
ESAIM: Probability and Statistics, Tome 5 (2001), pp. 119-140
We consider the problem of hypothesis testing within a monotone regression model. We propose a new test of the hypothesis : “” against the composite alternative : “” under the assumption that the true regression function is decreasing. The test statistic is based on the -distance between the isotonic estimator of and the function , since it is known that a properly centered and normalized version of this distance is asymptotically standard normally distributed under . We study the asymptotic power of the test under alternatives that converge to the null hypothesis.
Classification :
62G08, 62G10, 62G20
Keywords: nonparametric regression, isotonic estimator, goodness of fit test, asymptotic power
Keywords: nonparametric regression, isotonic estimator, goodness of fit test, asymptotic power
@article{PS_2001__5__119_0,
author = {Durot, C\'ecile and Tocquet, Anne-Sophie},
title = {Goodness of fit test for isotonic regression},
journal = {ESAIM: Probability and Statistics},
pages = {119--140},
year = {2001},
publisher = {EDP-Sciences},
volume = {5},
mrnumber = {1875667},
zbl = {0990.62041},
language = {en},
url = {http://geodesic.mathdoc.fr/item/PS_2001__5__119_0/}
}
Durot, Cécile; Tocquet, Anne-Sophie. Goodness of fit test for isotonic regression. ESAIM: Probability and Statistics, Tome 5 (2001), pp. 119-140. http://geodesic.mathdoc.fr/item/PS_2001__5__119_0/