On the asymptotic behaviour of unit-root tests in the presence of a Markov trend

被引：3

作者：

Psaradakis, Z ^{[1
]}

机构：

[1] Univ London, Birkbeck Coll, Sch Econ Math & Stat, London W1T 1LL, England

来源：

STATISTICS & PROBABILITY LETTERS | 2002年 / 57卷 / 01期

关键词：

Markov chain; non-stationarity; structural change; unit-root test;

D O I：

10.1016/S0167-7152(02)00063-9

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 examines the behaviour of unit-root tests for I(1) time series with drift which is subject to Markov regime changes. It is shown that the asymptotic null distributions of the popular Dickey-Fuller statistics are different from the standard asymptotic distributions obtained under a no-break assumption. Monte Carlo experiments are used to illustrate the finite-sample implications of the theoretical results. (C) 2002 Elsevier Science B.V. All rights reserved.

引用

页码：101 / 109

页数：9