This work evaluates the ex post forecasts precision of a set of short term models for the Consumer Price Index (CPI), the Producer Price Index (PPI) and the Gross Domestic Product (GDP) using a recent sample of Peruvian data. We seek to determine whether adding disaggregated information at the level of components improves the forecast precision of the those models. Short term projections constitute an integral part of any forecasting system since those are usually used as starting points for projections made using structural models. In that sense more precise short-term projections help to minimize the forecast errors of medium term models. We find that using disaggregated data improves the forecast precision of the CPI in the very short run but not that of the PPI and the GDP for the same time horizon, even when we use time-varying parameter dynamic models. Finally for forecast time horizons higher than 12 months the forecast precision of models for these three indexes can not be improved using disaggregated data.