Abstract
Real-world time series present patterns that change over time, making them difficult to forecast using only one forecasting model. Dynamic selection approaches have been highlighted in literature due to their accuracy and ability to model different local patterns. These approaches select one or more models from a pool (or ensemble) to forecast each test pattern. This selection is performed based on the pool’s performance in a Region of Competence (RoC), a set of samples most similar to a test pattern. The RoC definition, the pool creation, the number of selected models, and the function for combining the forecasts are critical issues for the dynamic selection approaches once their accuracy is closely related to them. This paper proposes a dynamic selection system based on a heterogeneous pool that performs a data-driven choice to determine: (i) the best RoC size, (ii) the set of the most competent forecasting models, and (iii) the most suitable combination function. The selection uses an RoC composed of the nearest antecedent windows to a test pattern. The proposal employs a heterogeneous pool comprising six forecasting models: Autoregressive Integrated Moving Average (ARIMA), Theta model, Support Vector Regression (SVR), Multilayer Perceptron (MLP), Extreme Learning Machine (ELM), and Long Short-Term Memory (LSTM). An experimental analysis performed using seven well-known data sets showed that the proposal overcame literature single and ensemble approaches, indicating that it is able to perform a better dynamic selection.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Sezer, O.B., Gudelek, M.U., Ozbayoglu, A.M.: Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Appl. Soft Comput. 90, 106181 (2020)
Kaushik, S., et al.: AI in healthcare: time-series forecasting using statistical, neural, and ensemble architectures. Front. Big Data 3, 4 (2020)
Karevan, Z., Suykens, J.A.: Transductive LSTM for time-series prediction: an application to weather forecasting. Neural Netw. 125, 1–9 (2020)
Pierros, I., Vlahavas, I.: Architecture-agnostic time-step boosting: a case study in short-term load forecasting. In: Pimenidis, E., Angelov, P., Jayne, C., Papaleonidas, A., Aydin, M. (eds.) Artificial Neural Networks and Machine Learning. ICANN 2022. LNCS, vol. 13531, pp. 556–568. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-15934-3_46
Elorrieta, F., Eyheramendy, S., Palma, W.: Discrete-time autoregressive model for unequally spaced time-series observations. Astron. Astrophys. 627, A120 (2019)
Hajirahimi, Z., Khashei, M.: Hybrid structures in time series modeling and forecasting: a review. Eng. Appl. Artif. Intell. 86, 83–106 (2019)
Tealab, A.: Time series forecasting using artificial neural networks methodologies: a systematic review. Future Comput. Inform. J. 3(2), 334–340 (2018)
Lim, B., Zohren, S.: Time-series forecasting with deep learning: a survey. Philos. Trans. R. Soc. A. Math. Phys. Eng. Sci. 379(2194), 20200209 (2021)
Cheng, C., et al.: Time series forecasting for nonlinear and nonstationary processes: a review and comparative study. IIE Trans. 47, 1053–1071 (2015)
Makridakis, S., Spiliotis, E., Assimakopoulos, V.: The M4 competition: 100,000 time series and 61 forecasting methods. Int. J. Forecast. 36, 54–74 (2019)
Silva, E.G., De Mattos Neto, P.S.G., Cavalcanti, G.D.C.: A dynamic predictor selection method based on recent temporal windows for time series forecasting. IEEE Access. 9, 108466–108479 (2021)
Neto, P.S.D.M., Firmino, P.R.A., Siqueira, H., Tadano, Y.D.S., Alves, T.A., De Oliveira, J.F.L., Marinho, M.H.D.N., Madeiro, F.: Neural-based ensembles for particulate matter forecasting. IEEE Access 9, 14470–14490 (2021)
Qi, M., Zhang, G.P.: An investigation of model selection criteria for neural network time series forecasting. Eur. J. Oper. Res. 132(3), 666–680 (2001)
Oliveira, M., Torgo, L.: Ensembles for time series forecasting. In: Asian Conference on Machine Learning, pp. 360–370. PMLR (2015)
Santos, D., et al.: Solar irradiance forecasting using dynamic ensemble selection. Appl. Sci. 12, 3510 (2022)
Saadallah, A., Priebe, F., Morik, K.: A drift-based dynamic ensemble members selection using clustering for time series forecasting. In: Brefeld, U., Fromont, E., Hotho, A., Knobbe, A., Maathuis, M., Robardet, C. (eds.) ECML PKDD 2019. LNCS (LNAI), vol. 11906, pp. 678–694. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-46150-8_40
Cerqueira, V., Torgo, L., Oliveira, M., Pfahringer, B.: Dynamic and heterogeneous ensembles for time series forecasting. In: 2017 IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 242–251 (2017)
Du, L., Gao, R., Suganthan, P., Wang, D.: Bayesian optimization based dynamic ensemble for time series forecasting. Inf. Sci. 591, 155–175 (2022)
Montero-Manso, P., Athanasopoulos, G., Hyndman, R.J., Talagala, T.S.: FFORMA: feature-based forecast model averaging. Int. J. Forecast. 36(1), 86–92 (2020)
Fu, Y., Wu, D., Boulet, B.: Reinforcement learning based dynamic model combination for time series forecasting. Proc. AAAI Conf. Artif. Intell. 36(6), 6639–6647 (2022)
Siami-Namini, S., Tavakoli, N., Siami Namin, A.: A comparison of ARIMA and LSTM in forecasting time series. In: 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 1394–1401 (2018)
de O. Santos Jùnior, D.S., de Oliveira, J.F., de Mattos Neto, P.S.: An intelligent hybridization of ARIMA with machine learning models for time series forecasting. Knowl. Based Syst. 175, 72–86 (2019)
Chen, J.L., Li, G.S.: Evaluation of support vector machine for estimation of solar radiation from measured meteorological variables. Theoret. Appl. Climatol. 115, 627–638 (2013)
Valente, J.M., Maldonado, S.: SVR-FFS: a novel forward feature selection approach for high-frequency time series forecasting using support vector regression. Expert Syst. Appl. 160, 113729 (2020)
Borghi, P.H., Zakordonets, O., Teixeira, J.P.: A COVID-19 time series forecasting model based on MLP ANN. Proc. Comput. Sci, 181, 940–947 (2021)
Song, G., Dai, Q.: A novel double deep ELMs ensemble system for time series forecasting. Knowl. Based Syst. 134, 31–49 (2017)
Thomakos, D.D., Nikolopoulos, K.: Forecasting multivariate time series with the theta method. J. Forecast. 34(3), 220–229 (2015)
Elsworth, S., Güttel, S.: Time Series Forecasting Using LSTM Networks: A Symbolic Approach (2020)
Hyndman, R., Khandakar, Y.: Automatic time series forecasting: the forecast package for R. J. Statist. Softw. 26, 1–22 (2008)
Assimakopoulos, V., Nikolopoulos, K.: The theta model: a decomposition approach to forecasting. Int. J. Forecast. 16, 521–530 (2000)
Fiorucci, J., Pellegrini, T., Louzada, F., Petropoulos, F.: The Optimised Theta Method (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Matos, G.M., de Mattos Neto, P.S.G. (2023). An Intelligent Dynamic Selection System Based on Nearest Temporal Windows for Time Series Forecasting. In: Iliadis, L., Papaleonidas, A., Angelov, P., Jayne, C. (eds) Artificial Neural Networks and Machine Learning – ICANN 2023. ICANN 2023. Lecture Notes in Computer Science, vol 14259. Springer, Cham. https://doi.org/10.1007/978-3-031-44223-0_14
Download citation
DOI: https://doi.org/10.1007/978-3-031-44223-0_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44222-3
Online ISBN: 978-3-031-44223-0
eBook Packages: Computer ScienceComputer Science (R0)