• Citer

Consistent estimation of the asymptotic covariance structure of multivariate serial correlation

par Melard, Guy ;Paesmans, Marianne ;Roy, Roch
Référence Journal of Time Series Analysis, 12, 4, page (351-361)
Publication Publié, 1991

Article révisé par les pairs

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Abstract. A method is proposed for estimating, in a consistent way, the asymptotic covariance structure of serial correlations for a multivariate second‐order stationary process. To obtain a consistent estimator of this structure, which is also of the non‐negative definite type, results relative to the scalar case are generalized. The method consists in weighting appropriately the elements of the sample autocorrelation matrices in a generalization of Bartlett's formula so that the estimator converges in probability. Several useful applications of the results of the paper are mentioned. Copyright © 1991, Wiley Blackwell. All rights reserved