A new test for the parametric form of the variance function in nonparametric regression

Loading...
Thumbnail Image

Date

2005-10-11T14:37:10Z

Journal Title

Journal ISSN

Volume Title

Publisher

Alternative Title(s)

Abstract

In the common nonparametric regression model the problem of testing for the parametric form of the conditional variance is considered. A stochastic process based on the difference between the empirical processes obtained from the standardized nonparametric residuals under the null hypothesis (of a specific parametric form of the variance function) and the alternative is introduced and its weak convergence established. This result is used for the construction of a Cramér von Mises type statistic for testing the parametric form of the conditional variance. The finite sample properties of a bootstrap version of this test are investigated by means of a simulation study. In particular the new procedure is compared with some of the currently available methods for this problem and its performance is illustrated by means of a data example.

Description

Table of contents

Keywords

Bootstrap, Kernel estimation, Nonparametric regression, Residual distribution, Testing heteroscedasticity, Testing homoscedasticity

Subjects based on RSWK

Citation