Testing heteroscedasticity in nonlinear and nonparametric regressions

被引:23
|
作者
Zheng, Xu [1 ]
机构
[1] Shanghai Jiao Tong Univ, Sch Econ, Antai Coll Econ & Management, Shanghai 200052, Peoples R China
来源
CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE | 2009年 / 37卷 / 02期
关键词
Heteroscedasticity; homoscedasticity; nonparametric tests; nonlinear regression; nonparametric regression; wild bootstrap; Monte Carlo simulation; interest rate volatility; PARAMETRIC FORM; VARIANCE FUNCTION; CONSISTENT TEST; MODELS; ROBUST; HETEROSKEDASTICITY; DIAGNOSTICS; VOLATILITY; GOODNESS; FIT;
D O I
10.1002/cjs.10020
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
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 37: 282-300; 2009 (C) 2009 Statistical Society of Canada
引用
收藏
页码:282 / 300
页数:19
相关论文
共 50 条