Datasets
Hierarchical Datasets
Here we host a collection of datasets used in previous hierarchical
research by Rangapuram et al. [2021], Olivares et al. [2023], and
Kamarthi et al. [2022]. The benchmark datasets utilized include
Australian Monthly Labour
(Labour
),
SF Bay Area daily Traffic
(Traffic
,
OldTraffic
),
Quarterly Australian Tourism Visits
(TourismSmall
),
Monthly Australian Tourism visits
(TourismLarge
,
OldTourismLarge
),
and daily Wikipedia article views
(Wiki2
).
Old datasets favor the original datasets with minimal target variable
preprocessing (Rangapuram et al. [2021], Olivares et al. [2023]),
while the remaining datasets follow PROFHIT experimental settings.
References
- Syama Sundar Rangapuram, Lucien D Werner, Konstantinos Benidis,
Pedro Mercado, Jan Gasthaus, Tim Januschowski. (2021). “End-to-End
Learning of Coherent Probabilistic Forecasts for Hierarchical Time
Series”. Proceedings of the 38th International Conference on Machine
Learning
(ICML).
- Kin G. Olivares, O. Nganba Meetei, Ruijun Ma, Rohan Reddy, Mengfei
Cao, Lee Dicker (2022).”Probabilistic Hierarchical Forecasting with
Deep Poisson Mixtures”. International Journal Forecasting, special
issue.
- Harshavardhan Kamarthi, Lingkai Kong, Alexander Rodriguez, Chao
Zhang, and B. Prakash. PROFHIT: Probabilistic robust forecasting for
hierarchical time-series. Computing Research Repository.URL
https://arxiv.org/abs/2206.07940.
source
Labour
Labour (freq:str='MS', horizon:int=8, papers_horizon:int=12, seasonality:int=12, test_size:int=125, tags_names:Tuple[str]=('Country', 'Country/Region', 'Country/Gender/Region', 'Country/Employment/Gender/Region'))
source
TourismLarge
TourismLarge (freq:str='MS', horizon:int=12, papers_horizon:int=12, seasonality:int=12, test_size:int=57, tags_names:Tuple[str]=('Country', 'Country/State', 'Country/State/Zone', 'Country/State/Zone/Region', 'Country/Purpose', 'Country/State/Purpose', 'Country/State/Zone/Purpose', 'Country/State/Zone/Region/Purpose'))
source
TourismSmall
TourismSmall (freq:str='Q', horizon:int=4, papers_horizon:int=4, seasonality:int=4, test_size:int=9, tags_names:Tuple[str]=('Country', 'Country/Purpose', 'Country/Purpose/State', 'Country/Purpose/State/CityNonCity'))
source
Traffic
Traffic (freq:str='D', horizon:int=14, papers_horizon:int=7, seasonality:int=7, test_size:int=91, tags_names:Tuple[str]=('Level1', 'Level2', 'Level3', 'Level4'))
source
Wiki2
Wiki2 (freq:str='D', horizon:int=14, papers_horizon:int=7, seasonality:int=7, test_size:int=91, tags_names:Tuple[str]=('Views', 'Views/Country', 'Views/Country/Access', 'Views/Country/Access/Agent', 'Views/Country/Access/Agent/Topic'))
source
OldTraffic
OldTraffic (freq:str='D', horizon:int=1, papers_horizon:int=1, seasonality:int=7, test_size:int=91, tags_names:Tuple[str]=('Level1', 'Level2', 'Level3', 'Level4'))
source
OldTourismLarge
OldTourismLarge (freq:str='MS', horizon:int=12, papers_horizon:int=12, seasonality:int=12, test_size:int=57, tags_names:Tuple[str]=('Country', 'Country/State', 'Country/State/Zone', 'Country/State/Zone/Region', 'Country/Purpose', 'Country/State/Purpose', 'Country/State/Zone/Purpose', 'Country/State/Zone/Region/Purpose'))
source
HierarchicalData
HierarchicalData ()
Initialize self. See help(type(self)) for accurate signature.