The assessment of some macroeconomic forecasts for Spain using aggregated accuracy indicators

Abstract

In this study, a new accuracy measure is introduced to solve an important practical problem in assessing the forecast accuracy: different predictions’ accuracy measures indicate different forecasts as the most accurate. The proposed accuracy measure, called the S indicator, is based on three dimensions of the forecasts accuracy: the summary statistics that take into account the size error, which were aggregated using the S1 indicator, the accuracy measures used in forecasts comparisons that are summarized using the S2 indicator and the directional and sign accuracy based on the S3 measure. For the Spanish inflation, the real GDP rate and the unemployment rate a comparative analysis of accuracy was made for predictions provided over the recent crisis period (2008-2013) by Bank of Spain, European Commission (EC), Organization for Economic Cooperation and Development (OECD) and International Monetary Fund (IMF). Own inflation rate and real GDP rate predictions based on a moving average model, and a auto-regressive moving average model, respectively, outperformed the experts’ anticipations.