Modelling of Economic Time Series and the Method of Cointegration

Abstract

The article is focused on the problem of modelling multidimensional non-stationary cointegrated processes. It is a modern method especially used for the description of multidimensional economic time series. The multidimensional process Yt is called cointegrated with the cointegrating vector ?, if the process ?´Yt is stationary or trend-stationary. For instance this property can be found in some series of economic indices which are predominantly non-stationary. Methods connected with estimates of cointegrating vectors and with a cointegration testing are applied to economic data. All the methods given were programmed in the computing environment MATLAB.