Munich Personal RePEc Archive

Joint Forecast Combination of Macroeconomic Aggregates and Their Components

Cobb, Marcus P A (2017): Joint Forecast Combination of Macroeconomic Aggregates and Their Components.

[img]
Preview
PDF
MPRA_paper_76556.pdf

Download (411kB) | Preview

Abstract

This paper presents a framework that extends forecast combination to include an aggregate and its components in the same process. This is done with the objective of increasing aggregate forecasting accuracy by using relevant disaggregate information and increasing disaggregate forecasting accuracy by providing a binding context for the component’s forecasts. The method relies on acknowledging that underlying a composite index is a well defined structure and its outcome is a fully consistent forecasting scenario. This is particularly relevant for people that are interested in certain components or that have to provide support for a particular aggregate assessment. In an empirical application with GDP data from France, Germany and the United Kingdom we find that the outcome of the combination method shows equal aggregate accuracy to that of equivalent traditional combination methods and a disaggregate accuracy similar or better to that of the best single models.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.