(44-5) 17 * << * >> * Russian * English * Content * All Issues

Hybrid approach for time series forecasting based on a penalty p-spline and evolutionary optimization
E.A. Kochegurova 1, E.Yu. Repina 1, O.B. Tsekhan 2

Tomsk Polytechnic University, Tomsk, Russia,
Yanka Kupala State University of Grodno, Grodno, Belarus

 PDF, 1225 kB

DOI: 10.18287/2412-6179-CO-667

Pages: 821-829.

Full text of article: Russian language.

Abstract:
In this work, a hybrid-forecasting model is proposed. The model includes a recursive penalty P-spline with parameters adaptation based on evolutionary optimization algorithms. In short-term forecasting, especially in real-time systems, the urgent task is to increase the forecast speed without compromising its quality. High forecasting speed has been achieved by an economical computational scheme of a recurrent P-spline with a shallow depth of prehistory. When combined with the adaptation of some parameters of the P-spline, such an approach allows you to control the forecast accuracy.

Keywords:
penalized spline, smoothing spline, digital filter, impulse infinite response (IIR filter), instrumental function, amplitude and phase-frequency response.

Citation:
Kochegurova EA, RepinaEY, Tsekhan OB. Hybrid approach for time series forecasting based on a penalty p-spline and evolutionary optimization. Computer Optics 2020; 44(5): 821-829. DOI: 10.18287/2412-6179-CO-667.

Acknowledgements:
The work was funded by the Russian Foundation for Basic Research under grant #18-07-01007.

References:

  1. Yin Y, Shang P. Forecasting traffic time series with multivariate predicting method. Appl Math Comput 2016; 291(1): 266-278.
  2. Agafonov AA, Yumaganov AS, Myasnikov VV. Big data analysis in a geoinformatic problem of short-term traffic flow forecasting based on a K nearest neighbors method [In Russian]. Computer Optics 2018; 42(6): 1101-1111. DOI: 10.18287/2412-6179-2018-42-6-1101-1111.
  3. Sbrana G, Silvestrini A, Venditti F. Short-term inflation forecasting: The M.E.T.A. approach. Int J Forecast 2017; 33: 1065-1081.
  4. Montgomery DC., Jennings CL., Kulahci M. Introduction to time series analysis and forecasting. Hoboken, New Jersey: John Wiley and Sons Inc; 2015.
  5. Wang H, Zhangc Q, Wud J, Panf S, Chene Y. Time series feature learning with labeled and unlabeled data.  Pattern Recognit 2019; 89: 55-66.
  6. Box J, Jenkins G. Time Series Analysis. Forecast and management. San Francisco: CA Holden-Day; 1976.
  7. Astakhova NN, Demidova LA, Nikulchev EV. Application of multi-purpose optimization for forecasting time series groups [In Russian]. Cybernetics and Programming 2016; 5: 175-190.
  8. Parmezan A, Lee H, Wu F. Metalearning for choosing feature selection algorithms in data mining: Proposal of a new framework. Expert Syst Appl 2017; 75: 1-24.
  9. Chuchueva AI. A model for predicting time series for a sample of maximum similarity [In Russian]. The thesis for the Candidate’s degree in Technical Sciences. Мoscow: 2012.
  10. Parmezan A, Souza V, Batistaa G. Evaluation of statistical and machine learning models for time series prediction: Identifying the state-of-the-art and the best conditions for the use of each model. Inf Sci 2019; 484: 302-337.
  11. Zaporozhtsev, I.F. Short-term forecasting of spatio-temporal variability of oceanographic characteristics by methods multidimensional time series analysis [In Russian]. The thesis for the Candidate’s degree in Technical Sciences. Murmansk: 2016.
  12. Demidova LA, Sokolova YS. Data classification based on the SVM algorithm and the k-nearest neighbor algorithm [In Russian]. Bulletin of the Ryazan State Radio Engineering University 2017; 62: 119-132.
  13. Hajirahimi Z, Khashei M. Hybrid structures in time series modeling and forecasting: A review. Eng Appl Artif Intell 2019; 86: 83-106.
  14. Lu C. Wavelet fuzzy neural networks for identification and predictive control of dynamic systems. IEEE Trans Ind Electron 2011; 58(7): 3046-3058.
  15. Averkin AN, Yarushev S. Hybrid approach for time series forecasting based on ANFIS and Fuzzy Cognitive Maps. Proc XXth IEEE Int Conf Soft Computing and Measurements (SCM 2017) 2017: 379-381.
  16. Chen MY, Chen BT A hybrid fuzzy time series model based on granular computing for stock price forecasting. Inf Sci 2015;  294: 227-241.
  17. Rafiei M, Niknam T, Khooban MH. Probabilistic forecasting of hourly electricity price by generalization of ELM for usage in improved wavelet neural network. IEEE Trans Ind Inf 2017; 13(1): 71-79.
  18. Zhang ML, Zhou ZH A k-nearest neighbor based algorithm for multi-label classification. Proc 1st IEEE Int Conf on Granular Computing 2005: 718-721.
  19. Chernoff K, Nielsen M. Weighting of the k-Nearest-Neighbors. Proc 20th IEEE Int Conf Pattern Recognit (ICPR) 2010: 666-669.
  20. Liu H, Zhang S. Noisy data elimination using mutual k-nearest neighbor for classification mining. J Syst Softw 2012; 85(5): 1067-1074.
  21. de Boor C. A practical guide to splines. New York: Springer-Verlag; 2001.
  22. Sharif S, Kamal S. Comparison of significant approaches of penalized spline regression (P-splines). Pakistan J Stat Oper Res 2018; 14(2): 289-303.
  23. Budakçı G, Dişibüyük Ç, Goldman R, Oruç H. Extending fundamental formulas from classical B-splines to quantum B-splines. J Comput Appl Math 2015; 282: 17-33.
  24. Eilers P, Marx B, Durbán M, Durbán M. Twenty years of P-splines. Statistics and Operations Research Transactions 2015; 39(2): 149-186.
  25. Yang L, Hong Y. Adaptive penalized splines for data smoothing. Comput Stat Data Anal 2017; 108: 70-83.
  26. Kochegurova EA, Gorokhova ES. Current Estimation of the Derivative of a Non-stationary Process Based on a Recurrent Smoothing Spline. Optoelectronics, Instrumentation and Data Processing 2016; 52(3): 280-285.
  27. Kochegurova EA, Kochegurov AI, Rozhkova NE. Frequency analysis of recurrence variational P-splines. Optoelectronics, Instrumentation and Data Processing 2017; 53(6): 591-598.
  28. Martín A, Lara-Cabrera R, Fuentes-Hurtado F, Naranjo V, Camacho D. EvoDeep: a new evolutionary approach for automatic Deep Neural Networks parametrization. J Parallel Distrib Comput 2018; 117: 180-191.
  29. Zhang KQ, Qu ZX, Dong YX, Lu HY, Leng WN, Wang JZ, Zhang WY. Research on a combined model based on linear and nonlinear features – A case study of wind speed forecasting. Renew Energy 2019; 130: 814-830.
  30. Panteleev AV, Metlitskaya DV, Aleshina EA. Methods of global optimization. Metaheuristic strategies and algorithms [In Russian]. Moscow: "VUZovskaya Kniga" Publisher; 2013.
  31. Gelfand IM, Tsetlin ML. The principle of nonlocal search in automatic optimization problems. Doklady Akademii Nauk SSSR 1961; 137(2): 295-298.
  32. Kovartsev AN, Popova-Kovartseva DA. On efficiently of parallel algorithms for global optimization of functions of several variables. Computer Optics 2011; 35(2): 256-261.
  33. Bergstra J, Bardenet R, Bengio Y, Kégl B. Algorithms for hyper-parameter optimization. Proc 25th Annual Conf on Neural Information Processing Systems (NIPS) 2011: 1-9.
  34. Menyailov ES. Review and analysis of existing modifications of genetic algorithms [In Russian]. Open Information and Computer Integrated Technologies 2015; 70: 244-254.
  35. Kochegurova EA, Khozhaev IY, Rybushkina SV. Some results of designing an IIR smoothing filter with p-splines. Int Rev Autom Control 2019; 12(4): 200-209.
  36. Shcherbakov MV, Brebels A, Shcherbakova NL, Tyukov AP, Janovsky TA, Kamaev VA. A survey of forecast error measures. World Appl Sci J 2013; 24(24): 171-176.
  37. Parmezan A, Batista G. ICMC-USP time series prediction repository. 2014. Source: <http://sites.labic.icmc.usp.br/aparmezan/publications/pdf/Repository_Parmezan_USP_2014_TSPR.pdf>.
  38. Babu C, Reddy BA moving-average filter based hybrid ARIMA–ANN model for forecasting time series data. Appl Soft Comput 2014; 23: 27-38.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: ko@smr.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20