Nonparametric comparison of several regression functions: exact and asymptotic theory
Munk, Axel ; Dette, Holger
Ann. Statist., Tome 26 (1998) no. 3, p. 2339-2368 / Harvested from Project Euclid
A new test is proposed for the comparison of two regression curves $f$ and $g$. We prove an asymptotic normal law under fixed alternatives which can be applied for power calculations, for constructing confidence regions and for testing precise hypotheses of a weighted $L_2$ distance between $f$ and $g$ . In particular, the problem of nonequal sample sizes is treated, which is related to a peculiar formula of the area between two step functions. These results are extended in various directions, such as the comparison of $k$ regression functions or the optimal allocation of the sample sizes when the total sample size is fixed. The proposed pivot statistic is not based on a nonparametric estimator of the regression curves and therefore does not require the specification of any smoothing parameter.
Publié le : 1998-12-14
Classification:  Comparison of regression curves,  nonparametric analysis of covariance,  heteroscedastic errors,  $k$-sample problem,  62G05,  62G10,  62G30,  62G07
@article{1024691474,
     author = {Munk, Axel and Dette, Holger},
     title = {Nonparametric comparison of several regression functions: exact
		 and asymptotic theory},
     journal = {Ann. Statist.},
     volume = {26},
     number = {3},
     year = {1998},
     pages = { 2339-2368},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1024691474}
}
Munk, Axel; Dette, Holger. Nonparametric comparison of several regression functions: exact
		 and asymptotic theory. Ann. Statist., Tome 26 (1998) no. 3, pp.  2339-2368. http://gdmltest.u-ga.fr/item/1024691474/