FORECASTING IN EASTERN EUROPEAN COUNTRIES

The thesis provides detailed empirical applications of two sets of forecasting methods, popular in the academic literature, using macroeconomic time series of three Eastern European countries (EECs): namely, the Czech republic, Hungary and Poland. The idea is to develop a natural extension to my previous studies, in particular those presented in Junicke (2017), where I applied Bayesian inference to produce an empirical estimation of a dynamic stochastic general equilibrium (DSGE) model for a small open economy. After discussing a survey of the literature on macroeconomics forecasting, which includes a discussion on the developments of methodology and accuracy measures, I first analyze the forecasts resulting from a model with theoretical grounds. Then, I turn to those resulting from a model with econometric foundations. My findings are twofold. First, it suggests that using different pure econometric models, allowing for parameters and covariance matrix to vary may improve the forecasting performance for EECs on average. Second, the DSGE models forecast better when trend inflation is explicitly taken into consideration.