Comparative analysis of forecasting methods of Ukrainian macroeconomic indexes

Abstract

Classical methods of shortterm forecasting of macroeconomic indicators of Ukraine, namely, the method of exponential smoothing (ES),autoregressive (AR), autoregressive with moving average (ARMA), method of data group handing (MDGH polynomial) were investigated and a comparative analysis of the accuracy of forecasting methods was performed in order to determine the most appropriate one. The analysis showed that with the increase of the training sample the values of the statistical parameters (coefficient of determination and the Durbin-Watson) for all models are near their ideal values; MDGH polynomial model based on linear and quadratic partial descriptions are the best in comparison with classical methods (ES, AR, ARMA) in forecasting macroeconomic indicators (CPI and the national GDP of the economy of Ukraine.