Modeling of Multivariate Systems using Vector Autoregression(VAR)

This study revolves around a group of variables which are coming under a single system. The multivariate systems are a group of individual components which works independently but makes the system as a whole. These variables tend to have an effect on the growth of each other within the system. This paper explores the relationship between such systems by developing a mathematical model. A Vector Autoregression (VAR) model is developed for a multivariate system, to analyze the relationship between the systems in the long and short run. The model is then used for forecasting the system. Using the Impulse Response analysis, a conclusion between the interactions of the variables is found. In the long term, the variables tend to be independent of the other variables as they become more stable. In short term, the variables have a dependence within them.