Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization

Shanika L Wickramasuriya, George Athanasopoulos, Rob J Hyndman

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Large collections of time series often have aggregation constraints due to product or geographical groupings. The forecasts for the most disaggregated series are usually required to add-up exactly to the forecasts of the aggregated series, a constraint we refer to as “coherence.” Forecast reconciliation is the process of adjusting forecasts to make them coherent.
The reconciliation algorithm proposed by Hyndman et al. is based on a generalized least squares estimator that requires an estimate of the covariance matrix of the coherency errors (i.e., the errors that arise due to incoherence). We show that this matrix is impossible to estimate in practice due to identifiability conditions.
We propose a new forecast reconciliation approach that incorporates the information from a full covariance matrix of forecast errors in obtaining a set of coherent forecasts. Our approach minimizes the mean squared error of the coherent forecasts across the entire collection of time series under the assumption of unbiasedness. The minimization problem has a closed-form solution. We make this solution scalable by providing a computationally efficient representation.
We evaluate the performance of the proposed method compared to alternative methods using a series of simulation designs which take into account various features of the collected time series. This is followed by an empirical application using Australian domestic tourism data. The results indicate that the proposed method works well with artificial and real data. Supplementary materials for this article are available online.
Original languageEnglish
Pages (from-to)804-819
Number of pages16
JournalJournal of the American Statistical Association
Volume114
Issue number526
DOIs
Publication statusPublished - 2019

Keywords

  • Aggregation
  • Australian tourism
  • Coherent forecasts
  • Contemporaneous error correlation
  • Forecast combinations
  • Spatial correlations

Cite this

@article{70d8ec0ee2ad4f5793487513371c8d5f,
title = "Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization",
abstract = "Large collections of time series often have aggregation constraints due to product or geographical groupings. The forecasts for the most disaggregated series are usually required to add-up exactly to the forecasts of the aggregated series, a constraint we refer to as “coherence.” Forecast reconciliation is the process of adjusting forecasts to make them coherent.The reconciliation algorithm proposed by Hyndman et al. is based on a generalized least squares estimator that requires an estimate of the covariance matrix of the coherency errors (i.e., the errors that arise due to incoherence). We show that this matrix is impossible to estimate in practice due to identifiability conditions.We propose a new forecast reconciliation approach that incorporates the information from a full covariance matrix of forecast errors in obtaining a set of coherent forecasts. Our approach minimizes the mean squared error of the coherent forecasts across the entire collection of time series under the assumption of unbiasedness. The minimization problem has a closed-form solution. We make this solution scalable by providing a computationally efficient representation.We evaluate the performance of the proposed method compared to alternative methods using a series of simulation designs which take into account various features of the collected time series. This is followed by an empirical application using Australian domestic tourism data. The results indicate that the proposed method works well with artificial and real data. Supplementary materials for this article are available online.",
keywords = "Aggregation, Australian tourism, Coherent forecasts, Contemporaneous error correlation, Forecast combinations, Spatial correlations",
author = "Wickramasuriya, {Shanika L} and George Athanasopoulos and Hyndman, {Rob J}",
year = "2019",
doi = "10.1080/01621459.2018.1448825",
language = "English",
volume = "114",
pages = "804--819",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor & Francis",
number = "526",

}

Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization. / Wickramasuriya, Shanika L; Athanasopoulos, George; Hyndman, Rob J.

In: Journal of the American Statistical Association, Vol. 114, No. 526, 2019, p. 804-819.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization

AU - Wickramasuriya, Shanika L

AU - Athanasopoulos, George

AU - Hyndman, Rob J

PY - 2019

Y1 - 2019

N2 - Large collections of time series often have aggregation constraints due to product or geographical groupings. The forecasts for the most disaggregated series are usually required to add-up exactly to the forecasts of the aggregated series, a constraint we refer to as “coherence.” Forecast reconciliation is the process of adjusting forecasts to make them coherent.The reconciliation algorithm proposed by Hyndman et al. is based on a generalized least squares estimator that requires an estimate of the covariance matrix of the coherency errors (i.e., the errors that arise due to incoherence). We show that this matrix is impossible to estimate in practice due to identifiability conditions.We propose a new forecast reconciliation approach that incorporates the information from a full covariance matrix of forecast errors in obtaining a set of coherent forecasts. Our approach minimizes the mean squared error of the coherent forecasts across the entire collection of time series under the assumption of unbiasedness. The minimization problem has a closed-form solution. We make this solution scalable by providing a computationally efficient representation.We evaluate the performance of the proposed method compared to alternative methods using a series of simulation designs which take into account various features of the collected time series. This is followed by an empirical application using Australian domestic tourism data. The results indicate that the proposed method works well with artificial and real data. Supplementary materials for this article are available online.

AB - Large collections of time series often have aggregation constraints due to product or geographical groupings. The forecasts for the most disaggregated series are usually required to add-up exactly to the forecasts of the aggregated series, a constraint we refer to as “coherence.” Forecast reconciliation is the process of adjusting forecasts to make them coherent.The reconciliation algorithm proposed by Hyndman et al. is based on a generalized least squares estimator that requires an estimate of the covariance matrix of the coherency errors (i.e., the errors that arise due to incoherence). We show that this matrix is impossible to estimate in practice due to identifiability conditions.We propose a new forecast reconciliation approach that incorporates the information from a full covariance matrix of forecast errors in obtaining a set of coherent forecasts. Our approach minimizes the mean squared error of the coherent forecasts across the entire collection of time series under the assumption of unbiasedness. The minimization problem has a closed-form solution. We make this solution scalable by providing a computationally efficient representation.We evaluate the performance of the proposed method compared to alternative methods using a series of simulation designs which take into account various features of the collected time series. This is followed by an empirical application using Australian domestic tourism data. The results indicate that the proposed method works well with artificial and real data. Supplementary materials for this article are available online.

KW - Aggregation

KW - Australian tourism

KW - Coherent forecasts

KW - Contemporaneous error correlation

KW - Forecast combinations

KW - Spatial correlations

UR - http://www.scopus.com/inward/record.url?scp=85055690481&partnerID=8YFLogxK

U2 - 10.1080/01621459.2018.1448825

DO - 10.1080/01621459.2018.1448825

M3 - Article

VL - 114

SP - 804

EP - 819

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 526

ER -