Research Repository

THE IMPACT OF PERSISTENT CYCLES ON ZERO FREQUENCY UNIT ROOT TESTS

Castro, Tomás del Barrio and Rodrigues, Paulo MM and Taylor, AM Robert (2013) 'THE IMPACT OF PERSISTENT CYCLES ON ZERO FREQUENCY UNIT ROOT TESTS.' Econometric Theory, 29 (6). pp. 1289-1313. ISSN 0266-4666

Full text not available from this repository.

Abstract

<jats:p>In this paper we investigate the impact of persistent (nonstationary or near nonstationary) cycles on the asymptotic and finite-sample properties of standard unit root tests. Results are presented for the augmented Dickey–Fuller (ADF) normalized bias and <jats:italic>t</jats:italic>-ratio-based tests (Dickey and Fuller, 1979, <jats:italic>Journal of the American Statistical Association</jats:italic> 745, 427–431; Said and Dickey, 1984; <jats:italic>Biometrika</jats:italic> 71, 599–607). the variance ratio unit root test of Breitung (2002, <jats:italic>Journal of Econometrics</jats:italic> 108, 343–363), and the <jats:italic>M</jats:italic> class of unit-root tests introduced by Stock (1999, in Engle and White (eds.), <jats:italic>A Festschrift in Honour of Clive W.J. Granger</jats:italic>) and Perron and Ng (1996, <jats:italic>Review of Economic Studies</jats:italic> 63, 435–463). We show that although the ADF statistics remain asymptotically pivotal (provided the test regression is properly augmented) in the presence of persistent cycles, this is not the case for the other statistics considered and show numerically that the size properties of the tests based on these statistics are too unreliable to be used in practice. We also show that the <jats:italic>t</jats:italic>-ratios associated with lags of the dependent variable of order greater than two in the ADF regression are asymptotically normally distributed. This is an important result as it implies that extant sequential methods (see Hall, 1994, <jats:italic>Journal of Business &amp; Economic Statistics</jats:italic> 17, 461–470; Ng and Perron, 1995, <jats:italic>Journal of the American Statistical Association</jats:italic> 90, 268–281) used to determine the order of augmentation in the ADF regression remain valid in the presence of persistent cycles.</jats:p>

Item Type: Article
Subjects: H Social Sciences > HG Finance
Divisions: Faculty of Social Sciences
Faculty of Social Sciences > Essex Business School
Faculty of Social Sciences > Essex Business School > Essex Finance Centre
SWORD Depositor: Elements
Depositing User: Elements
Date Deposited: 17 Nov 2014 12:58
Last Modified: 15 Jan 2022 00:47
URI: http://repository.essex.ac.uk/id/eprint/11247

Actions (login required)

View Item View Item