### Abstract

Many applications require forecasts for a hierarchy comprising a set of time scries along with aggregates of subsets of these scries. Hierarchical forecasting require not only good prediction accuracy at each level of the hierarchy, but also the coherency between different levels-the property that forecasts add up appropriately across the hierarchy. A fundamental limitation of prior research is the focus on forecasting the mean of each time series. We consider the situation where probabilistic forecasts are needed for each series in the hierarchy, and propose an algorithm to compute predictive distributions rather than mean forecasts only. Our algorithm has the advantage of synthesizing information from different levels in the hierarchy through a sparse forecast combination and a probabilistic hierarchical aggregation. We evaluate the accuracy of our forecasting algorithm on both simulated data and large-scale electricity smart meter data. The results show consistent performance gains compared to state-of-the art methods.

Original language | English |
---|---|

Title of host publication | Proceedings of the 34th International Conference on Machine Learning |

Editors | Doina Precup, Yee Whye Teh |

Place of Publication | Massachusetts USA |

Publisher | Proceedings of Machine Learning Research (PMLR) |

Pages | 3348-3357 |

Number of pages | 10 |

Publication status | Published - 1 Jan 2017 |

Event | International Conference on Machine Learning 2017 - International Convention Centre , Sydney , Australia Duration: 6 Aug 2017 → 11 Aug 2018 Conference number: 34th https://icml.cc/Conferences/2017 https://2017.icml.cc/ |

### Publication series

Name | Proceedings of Machine Learning Research |
---|---|

Volume | 70 |

ISSN (Print) | 1938-7228 |

### Conference

Conference | International Conference on Machine Learning 2017 |
---|---|

Abbreviated title | ICML 2017 |

Country | Australia |

City | Sydney |

Period | 6/08/17 → 11/08/18 |

Internet address |

### Cite this

*Proceedings of the 34th International Conference on Machine Learning*(pp. 3348-3357). (Proceedings of Machine Learning Research; Vol. 70). Massachusetts USA: Proceedings of Machine Learning Research (PMLR).

}

*Proceedings of the 34th International Conference on Machine Learning.*Proceedings of Machine Learning Research, vol. 70, Proceedings of Machine Learning Research (PMLR), Massachusetts USA, pp. 3348-3357, International Conference on Machine Learning 2017, Sydney , Australia, 6/08/17.

**Coherent probabilistic forecasts for hierarchical time series.** / Taieb, Souhaib Ben; Taylor, James W.; Hyndman, Rob J.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Coherent probabilistic forecasts for hierarchical time series

AU - Taieb, Souhaib Ben

AU - Taylor, James W.

AU - Hyndman, Rob J

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Many applications require forecasts for a hierarchy comprising a set of time scries along with aggregates of subsets of these scries. Hierarchical forecasting require not only good prediction accuracy at each level of the hierarchy, but also the coherency between different levels-the property that forecasts add up appropriately across the hierarchy. A fundamental limitation of prior research is the focus on forecasting the mean of each time series. We consider the situation where probabilistic forecasts are needed for each series in the hierarchy, and propose an algorithm to compute predictive distributions rather than mean forecasts only. Our algorithm has the advantage of synthesizing information from different levels in the hierarchy through a sparse forecast combination and a probabilistic hierarchical aggregation. We evaluate the accuracy of our forecasting algorithm on both simulated data and large-scale electricity smart meter data. The results show consistent performance gains compared to state-of-the art methods.

AB - Many applications require forecasts for a hierarchy comprising a set of time scries along with aggregates of subsets of these scries. Hierarchical forecasting require not only good prediction accuracy at each level of the hierarchy, but also the coherency between different levels-the property that forecasts add up appropriately across the hierarchy. A fundamental limitation of prior research is the focus on forecasting the mean of each time series. We consider the situation where probabilistic forecasts are needed for each series in the hierarchy, and propose an algorithm to compute predictive distributions rather than mean forecasts only. Our algorithm has the advantage of synthesizing information from different levels in the hierarchy through a sparse forecast combination and a probabilistic hierarchical aggregation. We evaluate the accuracy of our forecasting algorithm on both simulated data and large-scale electricity smart meter data. The results show consistent performance gains compared to state-of-the art methods.

UR - http://www.scopus.com/inward/record.url?scp=85048460620&partnerID=8YFLogxK

M3 - Conference Paper

T3 - Proceedings of Machine Learning Research

SP - 3348

EP - 3357

BT - Proceedings of the 34th International Conference on Machine Learning

A2 - Precup, Doina

A2 - Teh, Yee Whye

PB - Proceedings of Machine Learning Research (PMLR)

CY - Massachusetts USA

ER -