Synergy between an Improved Covariate Unit Root Test and Cross-sectionally Dependent Panel Data Unit Root Tests

Authors

  • Kaddour Hadri,

    1. Queen's University Management School, Queen's University Belfast
    Search for more papers by this author
  • Eiji Kurozumi,

    1. Hitotsubashi University
    Search for more papers by this author
  • Daisuke Yamazaki

    1. Hitotsubashi University
    Search for more papers by this author
    • We are grateful to two anonymous referees for useful comments. We are also grateful to participants at the 17th African Econometric Society meeting and seminar participants at National University of Singapore, Kyoto University, Queen's University Belfast and Hitotsubashi University. All errors are our responsibility. This research was supported by JSPS and BA under the Japan-Britain Research Cooperative Program, by the Seimei Foundation, the Ministry of Education, Culture, Sports, Science and Technology under Grants-in-Aid No. 23243038 and by the Global COE program of the Research Unit for Statistical and Empirical Analysis in Social Sciences, Hitotsubashi University.

Abstract

This paper proposes the use of an improved covariate unit root test which exploits the cross-sectional dependence information when the panel data null hypothesis of a unit root is rejected. More explicitly, to increase the power of the test, we suggest the utilization of more than one covariate and offer several ways to select the ‘best’ covariates from the set of potential covariates represented by the individuals in the panel. Employing our methods, we investigate the Prebish-Singer hypothesis for nine commodity prices. Our results show that this hypothesis holds for all but the price of petroleum.

Ancillary