Specification and stabiity of two equations vector autoregressive model for time series data analysis

Abstract

The most fertile areas of contemporary time series research concerns multiequation models. The vector autoregressive (VAR) model is the multivariate counter part of the univariate autoregressive model. A VAR model can be used in examining the relationships among a set of variables. The estimates of the parameters of the VAR model can be used for forecasting purposes. Forecasting with a VAR is a multivariable extension of forecasting using a simple autoregression. The VAR analysis tools such as Granger Causality tests, Impulse Response Analysis and Variance Decomposition models can be used in establishing relationships among economic variables; and in the formulation of structured economic models. In the present study a Two-Equations VAR model has been specified and stability conditions have been derived. Using Lag operators, the Two-Equations VAR model has been expressed in terms of Lag operators. This specification facilitates to study further inferential aspects of the VAR model for time series data analysis.