Примена на векторските авторегресивни модели во макроекономско моделирање

Abstract

This paper has two objectives. The first objective is to use a simple and complete approach in order to elaborate the concepts of nonstationarity, unit root tests, cointegration, vector autoregressive models, vector error correction models and their importance in modeling of macroeconomic time series. The second objective of this paper is to demonstrate a successful application of the previous concepts while analyzing and modeling the inflation in Republic of Macedonia. With the use of vector error correction model with block exogenous restrictions and cointegration of the selected variables, integrated of order one, models are created that confirm that the most important determinant of the inflation in Republic of Macedonia is the inflation in Euro Area. The group of other factors that influence the national inflation contains the variables such as the price of the oil on the world market, imports of goods, monetary aggregate M1 and inflation inertia. This paper contains explanation of the effect which the appointed factors have on the inflation, their intensity, direction of the effects and their duration. Two models are created and both models predict that the inflation rate will decrease in 2014. This decline is logical, since the models confirm the dependency of the domestic inflation from the inflation in the Euro area, where according to the latest reports, there is a real risk of deflation, and proper measures are taken for control of this difficult tendency. The author assumes that serious deflation is not a likely scenario for Republic of Macedonia, yet a decline in the domestic inflation in the months to follow is very plausible, as official statistics already confirm. The models identify the key determinants of the inflation and they forecast real estimations for the inflation rate, which leads to conclusion that the vector autoregressive models are powerful and exceptionally applicable models in macroeconomic research.