On Trend Breaks and Initial Condition in Unit Root Testing. / Skrobotov, Anton.
In: Journal of Time Series Econometrics, Vol. 10, No. 1, 0014, 20.12.2017.Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - On Trend Breaks and Initial Condition in Unit Root Testing
AU - Skrobotov, Anton
N1 - Publisher Copyright: © 2018 Walter de Gruyter GmbH, Berlin/Boston.
PY - 2017/12/20
Y1 - 2017/12/20
N2 - Recent approaches in unit root testing have taken into account the influences of initial conditions and data trend breaks via pre-testing and union of rejection testing strategies. This paper reviews existing methods, extends the methods of (Harvey, D. I., S. J. Leybourne, and A. M. R. Taylor. 2012b. "Unit Root Testing under a Local Break in Trend." Journal of Econometrics 167:140-167), and integrates these techniques to create a comprehensive testing strategy. Even when presented with nuisance parameters such as initial conditions and data breaks, this new strategy holds promising asymptotic and finite sample properties.
AB - Recent approaches in unit root testing have taken into account the influences of initial conditions and data trend breaks via pre-testing and union of rejection testing strategies. This paper reviews existing methods, extends the methods of (Harvey, D. I., S. J. Leybourne, and A. M. R. Taylor. 2012b. "Unit Root Testing under a Local Break in Trend." Journal of Econometrics 167:140-167), and integrates these techniques to create a comprehensive testing strategy. Even when presented with nuisance parameters such as initial conditions and data breaks, this new strategy holds promising asymptotic and finite sample properties.
KW - Asymptotic local power
KW - Infimum Dickey-Fuller tests
KW - Local trend break
KW - Pre-testing
KW - Union of rejection
KW - Unit root test
UR - http://www.scopus.com/inward/record.url?scp=85041034782&partnerID=8YFLogxK
U2 - 10.1515/jtse-2016-0014
DO - 10.1515/jtse-2016-0014
M3 - Article
AN - SCOPUS:85041034782
VL - 10
JO - Journal of Time Series Econometrics
JF - Journal of Time Series Econometrics
SN - 2194-6507
IS - 1
M1 - 0014
ER -
ID: 92711831