Fast computation of reconciled forecasts for hierarchical and grouped time series

It is shown that the least squares approach to reconciling hierarchical time series forecasts can be extended to much more general collections of time series with aggregation constraints. The constraints arise due to the need for forecasts of collections of time series to add up in the same way as t...

Full description

Saved in:
Bibliographic Details
Published inComputational statistics & data analysis Vol. 97; pp. 16 - 32
Main Authors Hyndman, Rob J., Lee, Alan J., Wang, Earo
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.05.2016
Subjects
Agglomeration
algorithms
Business
Collection
Combining forecasts
Computation
Data processing
Grouped time series
Hierarchical time series
least squares
Least squares method
Reconciling forecasts
Statistics
Time series
time series analysis
Weighted least squares
Reconciling forecasts
Weighted least squares
Combining forecasts
Hierarchical time series
Grouped time series
Online AccessGet full text

Cover

More Information
Summary:It is shown that the least squares approach to reconciling hierarchical time series forecasts can be extended to much more general collections of time series with aggregation constraints. The constraints arise due to the need for forecasts of collections of time series to add up in the same way as the observed time series. It is also shown that the computations involved can be handled efficiently by exploiting the structure of the associated design matrix, or by using sparse matrix routines. The proposed algorithms make forecast reconciliation feasible in business applications involving very large numbers of time series.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2015.11.007