Modelling and Predicting Foreign Direct Investments in Romania
Modelling and Predicting Foreign Direct Investments in Romania

Summary/Abstract: The main aim of this research is to propose some econometric models to predict the foreign direct investment (FDI) in Romania. For describing the evolution of this variable during 1990-2013, several econometric models were proposed: a moving average model, a model with lag and a vector-autoregression (VAR). All the proposed predictions are superior to naïve ones. The moving average model provided the best forecasts for FDI as percent of GDP during 2011-2013, the corresponding mean absolute percentage error (MAPE) being 6%. According to variance decomposition of VAR model, changes in real GDP rate hardly explain d_FDI variance in the short-run and long-run. The changes in the rate of real GDP account for 0.5% of d_FDI changes in the short-run and long-run.

• Details
• Contents