Generalized fixed-T panel unit root tests

Yiannis Karavias; Elias Tzavalis

doi:10.1111/sjos.12392

Generalized fixed-T panel unit root tests

Yiannis Karavias, Elias Tzavalis

Economics

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

219 Downloads (Pure)

Abstract

Panel data unit root tests which can be applied to data that do not have many time series observations are based on very restrictive error and deterministic component specification assumptions. In this paper we develop a new, doubly modified estimator, based on which we propose a panel unit root test that allows for multiple structural breaks, linear and non-linear trends, heteroscedasticity, serial correlation and error cross section heterogeneity, when the number of time series observations is finite. The test has the additional perk that it is invariant to the initial condition.

Original language	English
Pages (from-to)	1227-1251
Number of pages	25
Journal	Scandinavian Journal of Statistics
Volume	46
Issue number	4
DOIs	https://doi.org/10.1111/sjos.12392
Publication status	Published - Dec 2019

Keywords

fixed T
nonlinear trends
panel data
serial correlation
structural breaks
unit root

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1111/sjos.12392Licence: None: All rights reserved

Karavias_&_Tzavalis_Generalized_fixed_T_panel_Scandinavian_Journal_of_Statistics_2019
Checked for eligibility 18/02/2019 This is the peer reviewed version of the following article: Karavias, Y, Tzavalis, E. Generalized fixed‐T panel unit root tests. Scand J Statist. 2019; 1– 25 https://doi.org/10.1111/sjos.12392, which has been published in final form at https://doi.org/10.1111/sjos.12392. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.
Accepted author manuscript, 379 KBLicence: None: All rights reserved

https://onlinelibrary.wiley.com/doi/full/10.1111/sjos.12392Licence: None: All rights reserved

Cite this

@article{a22b4e70a98f4f20b0053f078945e612,

title = "Generalized fixed-T panel unit root tests",

abstract = "Panel data unit root tests which can be applied to data that do not have many time series observations are based on very restrictive error and deterministic component specification assumptions. In this paper we develop a new, doubly modified estimator, based on which we propose a panel unit root test that allows for multiple structural breaks, linear and non-linear trends, heteroscedasticity, serial correlation and error cross section heterogeneity, when the number of time series observations is finite. The test has the additional perk that it is invariant to the initial condition.",

keywords = "fixed T, nonlinear trends, panel data, serial correlation, structural breaks, unit root",

author = "Yiannis Karavias and Elias Tzavalis",

year = "2019",

month = dec,

doi = "10.1111/sjos.12392",

language = "English",

volume = "46",

pages = "1227--1251",

journal = "Scandinavian Journal of Statistics",

issn = "0303-6898",

publisher = "Wiley-Blackwell",

number = "4",

}

TY - JOUR

T1 - Generalized fixed-T panel unit root tests

AU - Karavias, Yiannis

AU - Tzavalis, Elias

PY - 2019/12

Y1 - 2019/12

N2 - Panel data unit root tests which can be applied to data that do not have many time series observations are based on very restrictive error and deterministic component specification assumptions. In this paper we develop a new, doubly modified estimator, based on which we propose a panel unit root test that allows for multiple structural breaks, linear and non-linear trends, heteroscedasticity, serial correlation and error cross section heterogeneity, when the number of time series observations is finite. The test has the additional perk that it is invariant to the initial condition.

AB - Panel data unit root tests which can be applied to data that do not have many time series observations are based on very restrictive error and deterministic component specification assumptions. In this paper we develop a new, doubly modified estimator, based on which we propose a panel unit root test that allows for multiple structural breaks, linear and non-linear trends, heteroscedasticity, serial correlation and error cross section heterogeneity, when the number of time series observations is finite. The test has the additional perk that it is invariant to the initial condition.

KW - fixed T

KW - nonlinear trends

KW - panel data

KW - serial correlation

KW - structural breaks

KW - unit root

UR - http://www.scopus.com/inward/record.url?scp=85062958861&partnerID=8YFLogxK

U2 - 10.1111/sjos.12392

DO - 10.1111/sjos.12392

M3 - Article

SN - 0303-6898

VL - 46

SP - 1227

EP - 1251

JO - Scandinavian Journal of Statistics

JF - Scandinavian Journal of Statistics

IS - 4

ER -

Generalized fixed-T panel unit root tests

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this