“…It is worth mentioning that in the case when the residual series has bivariate Gaussian distribution, then X 1 (t) and X 2 (t) for each t ∈ Z have one-dimensional Gaussian distributions and Y (t) has a variancegamma distribution with appropriate parameters (see Aroian et al, 1978). A detailed analysis related to the distribution of a product Gaussian random variables was presented, for instance, by Adamska et al (2021). In case when (Z 1 , Z 2 ) have the bivariate Student's t distribution, the components of the VAR(1) model X 1 (t) and X 2 (t) are not Student's t distributed.…”