Structural Identification of Quasi White Noise Linear Models
Anselmo Chaves Neto, Reinaldo Castro Souza, Isabella Andreczevski Chaves

Abstract
This paper shows that within the parametric admissible region of the Box and Jenkins ARMA(p,q) structures there is a sub-region where the classical identification procedure via ACF and PACF does not work properly. We define this sub-space the “Quasi White Noise region” and show how the bootstrap procedure can be used to improve the performance of the structural identification for series located in this particular region. A simulation study was carried out and a comparison between the traditional and the bootstrap procedures is presented.

Full Text: PDF