A panel cointegrating rank test with structural breaks and cross-sectional dependence

Research output: Journal contributionsJournal articlesResearchpeer-review


A new panel cointegrating rank test which allows for a linear time trend with breaks and cross-sectional dependence is proposed. The new correlation-augmented inverse normal (CAIN) test is based on a modification of the inverse normal method and combines the p-values of individual likelihood-ratio trace statistics by assuming that the number of breaks and break points are known. A Monte Carlo study demonstrates its robustness to cross-sectional dependence and its superior size and power properties compared to other meta-analytic tests used in practice. The test is applied to investigate the long-run relationship between regional house prices and personal income in the United States in view of the structural break introduced by the Global Financial Crisis.
Original languageEnglish
JournalEconometrics and Statistics
Pages (from-to)107-129
Number of pages23
Publication statusPublished - 01.01.2021

Bibliographical note

Funding Information:
Financial support by the German Research Foundation (DFG) through the project KA-3145/1-2 is gratefully acknowledged. The authors also thank two anonymous referees and the associate editor for many helpful comments and suggestions.

Publisher Copyright:
© 2020 The Author(s)

    Research areas

  • Economics - Panel cointegrating rank test, Structural breaks, Cross-sectional dependence, likelihood-ratio, Time trend