Abstract
The problem of selecting the optimum system of models for forecasting short-term railway traffic volumes is considered. The historical data is the daily volume of railway traffic between pairs of stations for different types of cargo. The given time series are highly volatile, noisy, and nonstationary. A system is proposed that selects the optimum superpositioning of forecasting models with respect to features of the historical data. A model of sliding averages, exponential and kernel-smoothing models, the ARIMA model, Croston’s method, and LSTM neural networks are considered as candidates for inclusion in superpositioning.
Similar content being viewed by others
References
E. V. Buryak, V. P. Kul’pina, A. V. Golyashev, A. A. Lobanova, “Dynamics of cargo transportation in Russia,” Byull. Sots.-Ekon. Krizisa Ross. 1 (4), 12–15 (2015).
L. Lijuan, H. Liu, J. Wu, et al., “A novel model for wind power forecasting based on Markov residual correction,” Renewable Energy Congress 6, 1–5 (2015).
D. A. Cohen and T. Z. Lys, “A note on analysts’ earnings forecast errors distribution,” J. Account. Econ. 36, 147–164 (2003).
M. L. Kan, Y. B. Lee, and W. C. Chen, “Apply grey prediction in the number of Tourist,” Genet. Evolut. Comput. 4, 481–484 (2010).
Yu. I. Zhuravlev, K. V. Rudakov, A. D. Korchagin, M. P. Kuznetsov, A. P. Motrenko, M.M. Stenina, and V. V. Strizhov, “Methods for forecasting time series on the example of railway cargo transportation,” Vestn. Ross. Akad. Nauk 86, 186–188 (2016).
B. Guha and G. Bandyopadhyay, “Gold price forecasting using ARIMA model,” J. Adv. Manage. Sci. 4, 117–121 (2016).
N. E. Golyandina, Method ‘Caterpillar’-SSA: Time Series Forecasting (Akademiya, Moscow, 2004) [in Russian].
N. Laptev, J. Yosinski, L. E. Li, et al., “Time-series extreme event forecasting with neural networks at Uber”, in Proceedings of the 34th International Conference on Machine Learning (Curran Assoc., New York, 2017), pp. 1–5.
P. S. Kalekar, “Time series forecasting using holt-winters exponential smoothing,” Kanwal Rekhi School of Information Technology (Leadstart, Mumbai, 2004), pp. 1–13.
L. Ralaivola and F. D’Alch-Buc, “Time series filtering, smoothing and learning using the kernel Kalman filter,” in Neural Networks, Proceedings of IEEE International Joint Conference (IEEE, New York, 2005), Vol. 3, pp. 1449–1454.
Railroad Transportation Time Series. URL: http://svn.code.sf.net/p/mlalgorithms/code/Group474/ Uvarov2017SuperpositionForecasting/data.
N. D. Uvarov, M. P. Kuznetsov, A. S. Mal’kova, K. V. Rudakov, and V. V. Strizhov, Appendix to article “Selection of superposition of models for railway freight forecasting.” URL: http://svn.code.sf.net/p/mlalgorithms/code/Group474/Uvarov2017SuperpositionForecasting/-doc/addition.pdf.
G. E. Box, G. M. Jenkins, G. C. Reinsel, et al., Time Series Analysis, Forecasting and Control (Wiley, New York, 1976).
Electricity Price (Germany). URL: http://svn.code.sf.net/p/mlalgorithms/code/Group474/Uvarov2017-SuperpositionForecasting/data/electricity_price_german/GermanSpotPrice.csv.
Electricity Consumption (Poland). URL: http://gdudek.el.pcz.pl/files/PL.xls.
Experiment Source code for “Selection of superposition of models for railway freight forecasting” (Python module and Jupyter-notebook). URL: http://svn.code.sf.net/p/mlalgorithms/code/Group474/-Uvarov2017SuperpositionForecasting/code/.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © N.D. Uvarov, M.P. Kuznetsov, A.S. Malkova, K.V. Rudakov, V.V. Strijov, 2018, published in Vestnik Moskovskogo Universiteta, Seriya 15: Vychislitel’naya Matematika i Kibernetika, 2018, No. 4, pp. 41–50.
About this article
Cite this article
Uvarov, N.D., Kuznetsov, M.P., Malkova, A.S. et al. Selecting the Superpositioning of Models for Railway Freight Forecasting. MoscowUniv.Comput.Math.Cybern. 42, 186–193 (2018). https://doi.org/10.3103/S027864191804009X
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S027864191804009X