Hybrid rank-based panel unit root tests

Ramon van den Akker; Bas J.M.  Werker; Bo Zhou

doi:10.1007/978-3-031-61853-6

Hybrid rank-based panel unit root tests

Ramon van den Akker
, Bas J.M. Werker
, Bo Zhou^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Scientific › peer-review

Abstract

We study optimality of unit root tests in a semiparametric panel setting with both large time-series and cross-sectional dimension. While, with finite cross-sectional dimension, the limit experiments are locally asymptotically Brownian functional (LABF), we find the classical locally asymptotically normal (LAN) structure with large cross-sectional dimension. This leads to an upper bound for the power of tests using sequential asymptotics. We propose a hybrid rank-based test, based on cross-sectional ranks per time unit, that is point optimal. We corroborate our theoretical results with a Monte Carlo study.

Original language	English
Title of host publication	Recent Advances in Econometrics and Statistics
Subtitle of host publication	Festschrift in Honour of Marc Hallin
Editors	Matteo Barigozzi, Siegfried Hörmann, Davy Paindaveine
Publisher	Springer
Pages	105-125
Number of pages	21
ISBN (Electronic)	978-3-031-61853-6
ISBN (Print)	978-3-031-61852-9, 978-3-031-61855-0
DOIs	https://doi.org/10.1007/978-3-031-61853-6
Publication status	Published - 29 Oct 2024

Access to Document

10.1007/978-3-031-61853-6

978-3-031-61853-6_6Final published version, 664 KB

Cite this

@inbook{841ed2b394964b7e9605e6a323f1752f,

title = "Hybrid rank-based panel unit root tests",

abstract = "We study optimality of unit root tests in a semiparametric panel setting with both large time-series and cross-sectional dimension. While, with finite cross-sectional dimension, the limit experiments are locally asymptotically Brownian functional (LABF), we find the classical locally asymptotically normal (LAN) structure with large cross-sectional dimension. This leads to an upper bound for the power of tests using sequential asymptotics. We propose a hybrid rank-based test, based on cross-sectional ranks per time unit, that is point optimal. We corroborate our theoretical results with a Monte Carlo study.",

author = "\{van den Akker\}, Ramon and Werker, \{Bas J.M.\} and Bo Zhou",

year = "2024",

month = oct,

day = "29",

doi = "10.1007/978-3-031-61853-6",

language = "English",

isbn = "978-3-031-61852-9",

pages = "105--125",

editor = "Matteo Barigozzi and Siegfried H{\"o}rmann and Davy Paindaveine",

booktitle = "Recent Advances in Econometrics and Statistics",

publisher = "Springer",

address = "Germany",

}

Hybrid rank-based panel unit root tests. / van den Akker, Ramon ; Werker, Bas J.M. ; Zhou, Bo.
Recent Advances in Econometrics and Statistics: Festschrift in Honour of Marc Hallin. ed. / Matteo Barigozzi; Siegfried Hörmann; Davy Paindaveine. Springer, 2024. p. 105-125.

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Scientific › peer-review

TY - CHAP

T1 - Hybrid rank-based panel unit root tests

AU - van den Akker, Ramon

AU - Werker, Bas J.M.

AU - Zhou, Bo

PY - 2024/10/29

Y1 - 2024/10/29

N2 - We study optimality of unit root tests in a semiparametric panel setting with both large time-series and cross-sectional dimension. While, with finite cross-sectional dimension, the limit experiments are locally asymptotically Brownian functional (LABF), we find the classical locally asymptotically normal (LAN) structure with large cross-sectional dimension. This leads to an upper bound for the power of tests using sequential asymptotics. We propose a hybrid rank-based test, based on cross-sectional ranks per time unit, that is point optimal. We corroborate our theoretical results with a Monte Carlo study.

AB - We study optimality of unit root tests in a semiparametric panel setting with both large time-series and cross-sectional dimension. While, with finite cross-sectional dimension, the limit experiments are locally asymptotically Brownian functional (LABF), we find the classical locally asymptotically normal (LAN) structure with large cross-sectional dimension. This leads to an upper bound for the power of tests using sequential asymptotics. We propose a hybrid rank-based test, based on cross-sectional ranks per time unit, that is point optimal. We corroborate our theoretical results with a Monte Carlo study.

U2 - 10.1007/978-3-031-61853-6

DO - 10.1007/978-3-031-61853-6

M3 - Chapter

SN - 978-3-031-61852-9

SN - 978-3-031-61855-0

SP - 105

EP - 125

BT - Recent Advances in Econometrics and Statistics

A2 - Barigozzi, Matteo

A2 - Hörmann, Siegfried

A2 - Paindaveine, Davy

PB - Springer

ER -

Hybrid rank-based panel unit root tests

Abstract

Access to Document

Fingerprint

Cite this