Forecasting Economic Aggregates Using Dynamic Component Grouping
MPRA Paper from University Library of Munich, Germany
Abstract In terms of aggregate accuracy, whether it is worth the effort of modelling a disaggregate process, instead of forecasting the aggregate directly, depends on the properties of the data. Forecasting the aggregate directly and forecasting each of the components separately, however, are not the only options. This paper develops a framework to forecast an aggregate that dynamically chooses groupings of components based on the properties of the data to benefit from both the advantages of aggregation and disaggregation. With this objective in mind, the dimension of the problem is reduced by selecting a subset of possible groupings through the use of agglomerative hierarchical clustering. The definitive forecast is then produced based on this subset. The results from an empirical application using CPI data for France, Germany and the UK suggest that the grouping methods can improve both aggregate and disaggregate accuracy.
Keywords: Forecasting economic aggregates; Bottom-up forecasting; Hierarchical forecasting; Hierarchical Clustering (search for similar items in EconPapers)
JEL-codes: C38 C53 E37 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-for, nep-mac and nep-ore
References: View references in EconPapers View complete reference list from CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
https://mpra.ub.uni-muenchen.de/81585/1/MPRA_paper_81585.pdf original version (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:pra:mprapa:81585
Access Statistics for this paper
More papers in MPRA Paper from University Library of Munich, Germany Ludwigstraße 33, D-80539 Munich, Germany. Contact information at EDIRC.
Bibliographic data for series maintained by Joachim Winter ().