A STRATEGY TO IMPROVE THE INFLATION RATE FORECASTS IN ROMANIA

The main goal of this research is to improve the degree of accuracy for inflation rate forecasts in Romania. The inflation was forecasted using a vectorial-autoregressive model. According to Granger test for causality, the relationship between the two variables is reciprocal. The inflation rate volatility is due mainly to the evolution of this indicator, the influence decreasing insignificantly in time, not descending under 96%. More than 87% of the variation in unemployment rate is explained by the own volatility for all lags. For the first lag the inflation is explained only by its evolution, the contribution of the unemployment rate to inflation variation being null. The inflation rate dynamic simulations (deterministic and stochastic) on the horizon 2011-2013 were more accurate than the predictions based on Dobrescu model. The combined forecasts proved to be a good strategy of improving the VAR forecasts and those based on Dobrescu model only if the dynamic and deterministic simulations were combined with Dobrescu’s anticipations on the horizon 2011-2013.