Forecasting the NN5 time series with hybrid models

JD Wichard - International Journal of Forecasting, 2011 - Elsevier
… We apply this approach to the NN5 time series competition data set. … All of the time series
show a weekly seasonality, as is indicated by the periodicity of the autocorrelation function in …

An accurate and fully-automated ensemble model for weekly time series forecasting

R Godahewa, C Bergmeir, GI Webb… - International Journal of …, 2023 - Elsevier
… there is no difference between “pure” weekly datasets and datasets aggregated into weekly
… all benchmarks and variants on the M4, NN5 and Kaggle web traffic datasets. For the M4 …

Forecast combinations of computational intelligence and linear models for the NN5 time series forecasting competition

RR Andrawis, AF Atiya, H El-Shishiny - International journal of forecasting, 2011 - Elsevier
… we participated in the NN5 time series competition. For the complete data set, our model was
… We then considered the weekly version of the time series (where each weekly data point is …

On utilizing self-organizing fuzzy neural networks for financial forecasts in the NN5 forecasting competition

D Coyle, G Prasad, TM McGinnity - The 2010 International Joint …, 2010 - ieeexplore.ieee.org
… autonomously cope with complex characteristics of datasets … The NN5 Competition The
objective of NN5 competition was … and daily demands vary weekly and seasonally, depending …

An averaging approach to individual time series employing econometric models: a case study on NN5 ATM transactions data

M Cedolin, M Erol Genevois - Kybernetes, 2022 - emerald.com
… ATM network by using the NN5 competition data set. Integrated … Especially, the NN5
competition data set helped researchers … They simulated weekly and monthly seasonality along …

[PDF][PDF] Neural Networks Ensembles Optimized via Simulated Annealing for NN5 Time Series Forecasting Competition

PJL Adeodato, AL Arnaud, GC Vasconcelos… - researchgate.net
… all the series had a weekly cyclic behavior later confirmed by … The test data set was then used
for producing the forecasts … the dependent to the independent data set and the quality itself, …

Meta-learning for time series forecasting in the NN GC1 competition

C Lemke, B Gabrys - International Conference on Fuzzy …, 2010 - ieeexplore.ieee.org
… Hence, as a contribution of this work, data of the NN3 and NN5 competitions have been
included in the meta-learning data set to extend the number of series for knowledge extraction. A …

Monash time series forecasting archive

R Godahewa, C Bergmeir, GI Webb… - arXiv preprint arXiv …, 2021 - arxiv.org
weekly version of this dataset is also … datasets with high frequencies such as weekly, daily,
and hourly show low seasonal strengths except for the NN5 weekly and NN5 daily datasets. …

[PDF][PDF] A strong baseline for weekly time series forecasting

R Godahewa, C Bergmeir, GI Webb… - arXiv preprint arXiv …, 2020 - academia.edu
weekly benchmark datasets by aggregating series with higher granularities. All the aggregated
weekly datasets … However, on the NN5, Kaggle web traffic, and solar datasets, FFORMA …

Can time series forecasting be automated? A benchmark and analysis

AT Sreedhara, J Vanschoren - arXiv preprint arXiv:2407.16445, 2024 - arxiv.org
… Note: Despite attempting to load the Kaggle daily dataset, … weekly dataset, which aggregates
the same time series into … • NN5 Weekly: Contains 111 weekly time series from the banking …