Premium
Testing heteroscedasticity in nonlinear and nonparametric regressions
Author(s) -
Zheng Xu
Publication year - 2009
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.1002/cjs.10020
Subject(s) - homoscedasticity , heteroscedasticity , nonparametric statistics , econometrics , null hypothesis , statistical hypothesis testing , monte carlo method , mathematics , statistics , nonparametric regression , volatility (finance)
This article presents new nonparametric tests for heteroscedasticity in nonlinear and nonparametric regression models. The tests have an asymptotic standard normal distribution under the null hypothesis of homoscedasticity and are robust against any form of heteroscedasticity. A Monte Carlo simulation with critical values obtained from the wild bootstrap procedure is provided to asses the finite sample performances of the tests. A real application of testing interest rate volatility functions illustrates the usefulness of the tests proposed. The Canadian Journal of Statistics © 2009 Statistical Society of Canada