THE ACCURACY OF UNEMPLOYMENT RATE FORECASTS IN ROMANIA AND THE ACTUAL ECONOMIC CRISIS

Abstract

In this study, the problem of forecasts accuracy is analysed on three different forecasting horizons: during the actual economic crisis, in few years before the crisis and on a large horizon. The accuracy of the forecasts made by European Commission, National Commission for Prognosis (NCP) and Institute for Economic Forecasting (IEF) for unemployment rate in Romania is assessed. The most accurate predictions on the forecasting horizons 2001-2011 and 2009-2011 were provided by IEF and the less accurate by NCP. These results were gotten using U1 Theil’s statistic and a new method that has not been used before in literature in this context. The multi-criteria ranking was applied to make a hierarchy of the institutions regarding the accuracy and five important accuracy measures were taken into account at the same time: mean errors, mean squared error, root mean squared error, U1 and U2 statistics of Theil. In few years before crisis (2006-2008) another hierarchy of institutions were gotten using the accuracy criterion: NCP, IEF and EC. The combined forecasts of institutions’ predictions are the best strategy to improve the forecasts accuracy on overall and before the crisis. During the economic crisis IEF provided the most accurate predictions, the combined forecasts being a good strategy of improving only the forecasts made by NCP and EC using inversely MSE scheme and equally weighted scheme. The assessment and improvement of forecasts accuracy have an important contribution in growing the quality of decisional process.

Keywords

• Forecasts
• Predictions
• Accuracy
• Multi-criteria ranking
• Combined forecasts
• Combining schemes