Mihaela BRATU


The GDP forecasting presents a particularity resulted from the fact that this macroeconomic indicator can be analyzed in its quality of aggregate. Therefore, the GDP can be predicted directly using an econometric model with lagged variables represented by the aggregate component. On the other hand, the same GDP can be predicted by aggregating the forecasts of its components. The aim of this study is to find out which strategy generates the most accurate one-step-ahead prediction and if combined forecasts can be a solution of improving the forecasts accuracy. Starting from the GDP one-year-ahead predictions made for 2009-2011 using the two strategies, measures of accuracy were calculated and the directly predicted GDP are better than those based on aggregating the components using constant and variable weights. Combined forecasts did not improve the accuracy of the predictions based on the mentioned strategies. This research is a good proof for putting the basis of considering the variables aggregation as an important source of uncertainty in forecasting.

Full text: PDF


source of uncertainty, forecasts, accuracy, disaggregation over variables, strategy of prediction, Diebold Mariano test, combined forecasts

JEL Codes

E21, E27


Armstrong, J. S. and Fildes, R., 1995. On the selection of Error Measures for Comparisons Among Forecasting Methods, Journal of Forecasting, 14(3), pp. 67.

Bates, J., and Granger, C. W. J., 1969. The Combination of Forecasts. Operations Research Quarterly, 20(4), pp. 17.

Clements, M. P. and Hendry, D. F . 1995. Forecasting in cointegrated systems. Journal of Applied Econometrics, 10(3), pp. 127.

Clements, M. P. and Hendry, D.F. 2010. Forecasting from Mis-specified Models in the Presence of Unanticipated Location Shifts, Department of Economics, Discussion Paper Series, 484(2), pp. 127.

Ericsson N. 2001. Forecast Uncertainty in Economic Modeling, MIT Press, Cambridge, pp. 68-92.

Hendry, D.F. and Clements, M.P. 2003. Evaluating a Model by Forecast Performance, Cambridge University Press, Cambridge, pp. 6.

Hendry, D. F. and Hubrich, K. 2006. Forecasting economic aggregates by disaggregates. Working Paper Series, European Central Bank, 589(2), pp. 8.

Hendry, D. F. and Hubrich, K. 2009. Combining disaggregate forecasts versus disaggregate information to forecast an aggregate. Journal of Business and Economic Statistics, 24(3), pp. 9.

Hubrich, K. 2005. Forecasting euro area inflation: Does aggregating forecasts by HICP component improve forecast accuracy?. International Journal of Forecasting, 21(1), pp. 119–136.

Hyndman, R. J. and Koehler, A.B. 2005. Another Look at Measures of Forecast Accuracy. Working Paper 13/05, Available at, [Accessed 22 May 2011].

Lanser, D. and Kranendonk, H. 2008. Investigating uncertainty in macroeconomic forecasts by stochastic simulation. CPB Discussion Paper, 112(4), pp. 3-5

Timmermann, A. 2006. Forecast Combinations, chap. 4, Handbook of Economic Forecasting. G. Elliott, C. Granger, and A. Timmermann, Elsevier, pp. 9.

Vega, M. 2003. Policy Makers Priors and Inflation Density Forecasts. Working Paper, Banco Central de la Reserva del Perú.

FRED, 2012. Data base. [online] Available at: [Accessed on September 2012].