Non-parametric testing of conditional variance functions in time series

被引:5
|
作者
Laïb, N [1 ]
机构
[1] Univ Paris 06, LSTA, F-93100 Montreuil, France
关键词
ARCH-models; autoregression; conditional variance; ergodic; goodness-of-fit;
D O I
10.1111/1467-842X.00298
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 paper proposes a non-parametric test for examining hypotheses about variance functions under stationarity and ergodicity conditions. Special cases of nonlinear time series models are studied, and it is found that under mild conditions the test is consistent. Its power is examined in a simulation study.
引用
收藏
页码:461 / 475
页数:15
相关论文
共 50 条