Estimation of pure autoregressive vector models for revenue series

Abstract

This paper aims at applying multivariate approach to Box and Jenkins univariate time series modeling to three vector series. General Autoregressive Vector Models with time varying coefficients are estimated. The first vector is a response vector, while others are predictor vectors. By matrix expansion each vector, whether response or predictor is a linear combination of other time varying vectors and itself. The models are linear and parameters are estimated by least squares method. The estimates obtained prove reality of the models estimated.

Global Jouranl of Mathematical Sciences Vol. 6 (1) 2007: pp. 31-38