Forecasting Unemployment Rate In Selected European Countries Using Smoothing Methods

Abstract

The aim of this paper is to select the most accurate<br> forecasting method for predicting the future values of the<br> unemployment rate in selected European countries. In order to do so,<br> several forecasting techniques adequate for forecasting time series<br> with trend component, were selected, namely: double exponential<br> smoothing (also known as Holt`s method) and Holt-Winters` method<br> which accounts for trend and seasonality. The results of the empirical<br> analysis showed that the optimal model for forecasting<br> unemployment rate in Greece was Holt-Winters` additive method. In<br> the case of Spain, according to MAPE, the optimal model was double<br> exponential smoothing model. Furthermore, for Croatia and Italy the<br> best forecasting model for unemployment rate was Holt-Winters`<br> multiplicative model, whereas in the case of Portugal the best model<br> to forecast unemployment rate was Double exponential smoothing<br> model. Our findings are in line with European Commission<br> unemployment rate estimates.