4.7 Further reading

  • The idea of using STL for features originated with Wang et al. (2006).
  • The features provided by the feasts package were motivated by their use in Hyndman et al. (2015) and Kang et al. (2017).
  • The exploration of a set of time series using principal components on a large collection of features was proposed by Kang et al. (2017).


Hyndman, R. J., Wang, E., & Laptev, N. (2015). Large-scale unusual time series detection. Proceedings of the IEEE International Conference on Data Mining, 1616–1619. [DOI]
Kang, Y., Hyndman, R. J., & Smith-Miles, K. (2017). Visualising forecasting algorithm performance using time series instance spaces. International Journal of Forecasting, 33(2), 345–358. [DOI]
Wang, X., Smith, K. A., & Hyndman, R. J. (2006). Characteristic-based clustering for time series data. Data Mining and Knowledge Discovery, 13(3), 335–364. [DOI]