Forecasting aggregate and disaggregates with common features

This paper focuses on providing consistent forecasts for an aggregate economic indicator, such as a consumer price index, and all its components, and on showing that the indirect forecast of the aggregate is at least as accurate as the direct forecast. The procedure developed is a disaggregated approach based on single-equation models for the components, which take into account the stable features as common trend and common serial correlation that some components have in common. Our procedure starts by classifying a large number of components based on restrictions from common features. The result of this classification is a disaggregation map, which may also be useful in applying dynamic factors, defining intermediate aggregates or formulating models with unobserved components. We apply the procedure to forecast inflation in the Euro area, the UK and the US. Our forecasts are significantly more accurate than a direct forecast of the aggregate and other indirect forecasts.