Improving the forecasting performance of temporal hierarchies

• Published in: PloS one Vol. 14; no. 10; p. e0223422
• Main Authors: Spiliotis, Evangelos, Petropoulos, Fotios, Assimakopoulos, Vassilios
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 03.10.2019; Public Library of Science (PLoS)
• Subjects: Accuracy; Algorithms; Bias; Biology and Life Sciences; Competition; Computer and Information Sciences; Computer engineering; Data smoothing; Displays (Marketing); Earth Sciences; Forecasting; Hierarchies; Humans; Inferential statistics; Management science; Methods; Models, Statistical; Multilevel analysis; Performance evaluation; Physical Sciences; Reproducibility of Results; Research and Analysis Methods; Seasons; Shrinkage; Social Sciences; Spatio-Temporal Analysis; Time series; Greece

Temporal hierarchies have been widely used during the past few years as they are capable to provide more accurate coherent forecasts at different planning horizons. However, they still display some limitations, being mainly subject to the forecasting methods used for generating the base forecasts an...