IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2201.00426.html
   My bibliography  Save this paper

Deep Learning and Linear Programming for Automated Ensemble Forecasting and Interpretation

Author

Listed:
  • Lars Lien Ankile
  • Kjartan Krange

Abstract

This paper presents an ensemble forecasting method that shows strong results on the M4 Competition dataset by decreasing feature and model selection assumptions, termed DONUT (DO Not UTilize human beliefs). Our assumption reductions, primarily consisting of auto-generated features and a more diverse model pool for the ensemble, significantly outperform the statistical, feature-based ensemble method FFORMA by Montero-Manso et al. (2020). We also investigate feature extraction with a Long Short-term Memory Network (LSTM) Autoencoder and find that such features contain crucial information not captured by standard statistical feature approaches. The ensemble weighting model uses LSTM and statistical features to combine the models accurately. The analysis of feature importance and interaction shows a slight superiority for LSTM features over the statistical ones alone. Clustering analysis shows that essential LSTM features differ from most statistical features and each other. We also find that increasing the solution space of the weighting model by augmenting the ensemble with new models is something the weighting model learns to use, thus explaining part of the accuracy gains. Moreover, we present a formal ex-post-facto analysis of an optimal combination and selection for ensembles, quantifying differences through linear optimization on the M4 dataset. Our findings indicate that classical statistical time series features, such as trend and seasonality, alone do not capture all relevant information for forecasting a time series. On the contrary, our novel LSTM features contain significantly more predictive power than the statistical ones alone, but combining the two feature sets proved the best in practice.

Suggested Citation

  • Lars Lien Ankile & Kjartan Krange, 2022. "Deep Learning and Linear Programming for Automated Ensemble Forecasting and Interpretation," Papers 2201.00426, arXiv.org, revised Nov 2022.
  • Handle: RePEc:arx:papers:2201.00426
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2201.00426
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Makridakis, Spyros & Hibon, Michele, 2000. "The M3-Competition: results, conclusions and implications," International Journal of Forecasting, Elsevier, vol. 16(4), pages 451-476.
    2. repec:cup:cbooks:9781108422536 is not listed on IDEAS
    3. Fildes, Robert & Hibon, Michele & Makridakis, Spyros & Meade, Nigel, 1998. "Generalising about univariate forecasting methods: further empirical evidence," International Journal of Forecasting, Elsevier, vol. 14(3), pages 339-358, September.
    4. Smyl, Slawek, 2020. "A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting," International Journal of Forecasting, Elsevier, vol. 36(1), pages 75-85.
    5. Clemen, Robert T., 1989. "Combining forecasts: A review and annotated bibliography," International Journal of Forecasting, Elsevier, vol. 5(4), pages 559-583.
    6. Kwiatkowski, Denis & Phillips, Peter C. B. & Schmidt, Peter & Shin, Yongcheol, 1992. "Testing the null hypothesis of stationarity against the alternative of a unit root : How sure are we that economic time series have a unit root?," Journal of Econometrics, Elsevier, vol. 54(1-3), pages 159-178.
    7. Rainer Schöbel & Jianwei Zhu, 1999. "Stochastic Volatility With an Ornstein–Uhlenbeck Process: An Extension," Review of Finance, European Finance Association, vol. 3(1), pages 23-46.
    8. Bessembinder, Hendrik, et al, 1995. "Mean Reversion in Equilibrium Asset Prices: Evidence from the Futures Term Structure," Journal of Finance, American Finance Association, vol. 50(1), pages 361-375, March.
    9. Capistrán, Carlos & Timmermann, Allan, 2009. "Forecast Combination With Entry and Exit of Experts," Journal of Business & Economic Statistics, American Statistical Association, vol. 27(4), pages 428-440.
    10. Montero-Manso, Pablo & Athanasopoulos, George & Hyndman, Rob J. & Talagala, Thiyanga S., 2020. "FFORMA: Feature-based forecast model averaging," International Journal of Forecasting, Elsevier, vol. 36(1), pages 86-92.
    11. Timmermann, Allan, 2006. "Forecast Combinations," Handbook of Economic Forecasting, in: G. Elliott & C. Granger & A. Timmermann (ed.), Handbook of Economic Forecasting, edition 1, volume 1, chapter 4, pages 135-196, Elsevier.
    12. Makridakis, Spyros & Chatfield, Chris & Hibon, Michele & Lawrence, Michael & Mills, Terence & Ord, Keith & Simmons, LeRoy F., 1993. "The M2-competition: A real-time judgmentally based forecasting study," International Journal of Forecasting, Elsevier, vol. 9(1), pages 5-22, April.
    13. Armstrong, J. Scott & Collopy, Fred, 1992. "Error measures for generalizing about forecasting methods: Empirical comparisons," International Journal of Forecasting, Elsevier, vol. 8(1), pages 69-80, June.
    14. Brooks,Chris, 2019. "Introductory Econometrics for Finance," Cambridge Books, Cambridge University Press, number 9781108436823.
    15. G. Elliott & C. Granger & A. Timmermann (ed.), 2006. "Handbook of Economic Forecasting," Handbook of Economic Forecasting, Elsevier, edition 1, volume 1, number 1.
    Full references (including those not matched with items on IDEAS)

    Citations

    Blog mentions

    As found by EconAcademics.org, the blog aggregator for Economics research:
    1. Deep Recurrent Neural Nets with Long Short Term Memory
      by Francis Diebold in No Hesitations on 2022-02-12 13:01:00

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Petropoulos, Fotios & Apiletti, Daniele & Assimakopoulos, Vassilios & Babai, Mohamed Zied & Barrow, Devon K. & Ben Taieb, Souhaib & Bergmeir, Christoph & Bessa, Ricardo J. & Bijak, Jakub & Boylan, Joh, 2022. "Forecasting: theory and practice," International Journal of Forecasting, Elsevier, vol. 38(3), pages 705-871.
      • Fotios Petropoulos & Daniele Apiletti & Vassilios Assimakopoulos & Mohamed Zied Babai & Devon K. Barrow & Souhaib Ben Taieb & Christoph Bergmeir & Ricardo J. Bessa & Jakub Bijak & John E. Boylan & Jet, 2020. "Forecasting: theory and practice," Papers 2012.03854, arXiv.org, revised Jan 2022.
    2. Li, Li & Kang, Yanfei & Li, Feng, 2023. "Bayesian forecast combination using time-varying features," International Journal of Forecasting, Elsevier, vol. 39(3), pages 1287-1302.
    3. Kang, Yanfei & Spiliotis, Evangelos & Petropoulos, Fotios & Athiniotis, Nikolaos & Li, Feng & Assimakopoulos, Vassilios, 2021. "Déjà vu: A data-centric forecasting approach through time series cross-similarity," Journal of Business Research, Elsevier, vol. 132(C), pages 719-731.
    4. Karsten R. Gerdrup & Anne Sofie Jore & Christie Smith & Leif Anders Thorsrud, 2009. "Evaluating ensemble density combination - forecasting GDP and inflation," Working Paper 2009/19, Norges Bank.
    5. Xiaojie Xu, 2020. "Corn Cash Price Forecasting," American Journal of Agricultural Economics, John Wiley & Sons, vol. 102(4), pages 1297-1320, August.
    6. Makridakis, Spyros & Hyndman, Rob J. & Petropoulos, Fotios, 2020. "Forecasting in social settings: The state of the art," International Journal of Forecasting, Elsevier, vol. 36(1), pages 15-28.
    7. Makridakis, Spyros & Spiliotis, Evangelos & Assimakopoulos, Vassilios, 2020. "The M4 Competition: 100,000 time series and 61 forecasting methods," International Journal of Forecasting, Elsevier, vol. 36(1), pages 54-74.
    8. Theodosiou, Marina, 2011. "Forecasting monthly and quarterly time series using STL decomposition," International Journal of Forecasting, Elsevier, vol. 27(4), pages 1178-1195, October.
    9. Godahewa, Rakshitha & Bergmeir, Christoph & Webb, Geoffrey I. & Montero-Manso, Pablo, 2023. "An accurate and fully-automated ensemble model for weekly time series forecasting," International Journal of Forecasting, Elsevier, vol. 39(2), pages 641-658.
    10. Armstrong, J. Scott & Green, Kesten C. & Graefe, Andreas, 2015. "Golden rule of forecasting: Be conservative," Journal of Business Research, Elsevier, vol. 68(8), pages 1717-1731.
    11. Paroissien, Emmanuel, 2020. "Forecasting bulk prices of Bordeaux wines using leading indicators," International Journal of Forecasting, Elsevier, vol. 36(2), pages 292-309.
    12. Vitor Azevedo & Georg Sebastian Kaiser & Sebastian Mueller, 2023. "Stock market anomalies and machine learning across the globe," Journal of Asset Management, Palgrave Macmillan, vol. 24(5), pages 419-441, September.
    13. Athanasopoulos, George & Hyndman, Rob J. & Kourentzes, Nikolaos & Petropoulos, Fotios, 2017. "Forecasting with temporal hierarchies," European Journal of Operational Research, Elsevier, vol. 262(1), pages 60-74.
    14. Weron, Rafał, 2014. "Electricity price forecasting: A review of the state-of-the-art with a look into the future," International Journal of Forecasting, Elsevier, vol. 30(4), pages 1030-1081.
    15. Wilson, Kevin J., 2017. "An investigation of dependence in expert judgement studies with multiple experts," International Journal of Forecasting, Elsevier, vol. 33(1), pages 325-336.
    16. Cheng, Gang & Yang, Yuhong, 2015. "Forecast combination with outlier protection," International Journal of Forecasting, Elsevier, vol. 31(2), pages 223-237.
    17. Timmermann, Allan, 2006. "Forecast Combinations," Handbook of Economic Forecasting, in: G. Elliott & C. Granger & A. Timmermann (ed.), Handbook of Economic Forecasting, edition 1, volume 1, chapter 4, pages 135-196, Elsevier.
    18. Wang, Xiaoqian & Hyndman, Rob J. & Li, Feng & Kang, Yanfei, 2023. "Forecast combinations: An over 50-year review," International Journal of Forecasting, Elsevier, vol. 39(4), pages 1518-1547.
    19. Davide De Gaetano, 2017. "Forecasting With Garch Models Under Structural Breaks: An Approach Based On Combinations Across Estimation Windows," Departmental Working Papers of Economics - University 'Roma Tre' 0219, Department of Economics - University Roma Tre.
    20. Adam Elbourne & Henk Kranendonk & Rob Luginbuhl & Bert Smid & Martin Vromans, 2008. "Evaluating CPB's published GDP growth forecasts; a comparison with individual and pooled VAR based forecasts," CPB Document 172, CPB Netherlands Bureau for Economic Policy Analysis.

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2201.00426. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.