Spurious rejection using recursive, rolling and sequential tests in the presence of a break under the null

  1. Badillo, R. 1
  2. Belaire, J. 2
  3. Reverte, C. 1
  1. 1 Universidad Politécnica de Cartagena
    info

    Universidad Politécnica de Cartagena

    Cartagena, España

    ROR https://ror.org/02k5kx966

  2. 2 Universitat de València
    info

    Universitat de València

    Valencia, España

    ROR https://ror.org/043nxc105

Actas:
XXXIII Simposio de Análisis Económico

Año de publicación: 2008

Tipo: Aportación congreso

GOOGLE SCHOLAR lock_openAcceso abierto editor

Resumen

Leybourne et al. (1998) examine the behaviour of conventional Dickey-Fuller tests in presence of a structural break under the null. They show that, if the break occurs early in the series a ‘converse Perron phenomenon’ is occurred. In this paper, we analyse whether their results change when the structural break is identified endogenously, that is, if the break point is related to the data. Applying recursive, rolling and sequential Dickey-Fuller type tests that control endogenously for structural breaks, we find no evidence of this phenomenon when using the sequential procedure.