This paper extents Karanasos (1999a) results for the n Component GARCH(1,1) and the two Component GARCH(2,2) models and it further examines the n Component GARCH(n,n) model. In particular, we present the GARCH(n^2;n^2) representation of the aggregate variance and we give the condition for the existence of the fourth moment of the errors. In addition, we use the canonical factorization of the autocovariance generating function for the univariate ARMA representations of the component variances, the aggregate variance and the squared errors to obtain their autocovariances and cross covariances. Finally, we illustrate our general results giving three examples: the three component GARCH(1,1), the two component GARCH(2,2) and the three component GARCH(2,2) models.
