Meta-learning how to forecast time series

 

Features of time series are useful in identifying suitable models for forecasting. We present a general framework, labelled FFORMS (Feature-based FORecast Model Selection), which selects forecast models based on features calculated from each time series. The FFORMS framework builds a mapping that relates the features of a time series to the “best” forecast model using a classification algorithm such as a random forest. The framework is evaluated using time series from the M-forecasting competitions and is shown to yield forecasts that are almost as accurate as state-of-the-art methods, but are much faster to compute. We use model-agnostic machine learning interpretability methods to explore the results and to study what types of time series are best suited to each forecasting model.

Associated R package: seer

Shiny app: https://thiyangt.github.io/fformsviz/fforms.html

Cite

@article{fforms,
  author  = {Thiyanga S Talagala and Rob J Hyndman and George Athanasopoulos},
  doi     = {10.1002/for.2963},
  journal = {J Forecasting},
  number  = 6,
  pages   = {1476-1501},
  title   = {Meta-learning how to forecast time series},
  url     = {http://robjhyndman.com/publications/fforms/},
  volume  = 42,
  year    = 2023
}