Many economic variables are often used in logarithms for forecasting and estimation analysis, as this transformation is considered to create a more homogeneous variance. The conditions under which the log transformation can help the researcher to produce more accurate forecasts are examined in previous studies in the literature. On the other hand, forecasting macroeconomic variables across a (large) number of countries/groups is also a difficult but standard task for economic analysts. A lot of work has been done about aggregated data and the way we can obtain optimal forecasts for the aggregate. However, the majority of those studies refer to linear transformations, and given that the logarithmic transformation is a nonlinear one, the effect of using logs in forecasting aggregated variables is not clear and needs further investigation. The aim is to investigate the combined effect of aggregation and the log transformation on forecasting. The study initially describes the alternatives approaches that can be followed to obtain forecasts for the variable of interest when this variable is an aggregated process. Each approach generates a different predictor. Then, we investigate the relative forecasting accuracy of each predictor by means of Monte Carlo simulations. Finally, a variety of empirical applications are carried out using aggregated economic variables.