Keywords
Deep learning
Fuzzy time series
Universe of discourse
Non-stationary time series
Variable-length interval partitioning
How to Cite
Abstract
Non-stationary time series prediction is challenging due to its dynamic and complex nature. Fuzzy time series models offer a promising solution for forecasting such data, but a key challenge lies in partitioning the universe of discourse, which significantly impacts forecasting accuracy. Traditional fuzzy time series models often use equal-length interval partitioning, which is more suited for stationary data and limits their adaptability to non-stationary time series. This paper introduces a novel variable-length interval partitioning method designed specifically for non-stationary time series. The developed method combines a Long Short-Term Memory (LSTM) Autoencoder with K-means clustering, enabling dynamic, data-driven partitioning that adapts to the changing characteristics of the data. The LSTM Autoencoder encodes the time series, which is clustered using K-means, and intervals are defined based on cluster centers. Furthermore, the Variable Length Interval Partitioning-based Fuzzy Time Series model (VLIFTS) is developed by incorporating this partitioning method and the concepts of Markov chain and transition probability matrix. In this model, fuzzy sets are viewed as states of a Markov chain, and transition probabilities are used in the forecasting phase. The model is validated on stock market indices Nifty 50, NASDAQ, S&P 500, and Dow Jones. Stationarity and heteroscedasticity are tested using Augmented Dickey-Fuller (ADF) and Levene's tests respectively. Statistical forecast accuracy metrics Root Mean Squared Error (RMSE) and Mean Absolute Percent Error (MAPE) show that VLIFTS significantly improves forecasting accuracy over traditional models. This hybrid approach enhances fuzzy time series modelling and can be applied to various non-stationary time series forecasting problems.
References
Zadeh LA. Fuzzy sets. Inf Control. 1965; 8(3): 338-53. https://doi.org/10.1016/S0019-9958(65)90241-X
Huarng K. Effective lengths of intervals to improve forecasting in fuzzy time series. Fuzzy Sets Syst. 2001; 123(3): 387-94. https://doi.org/10.1016/S0165-0114(00)00057-9
Chen SM. Forecasting enrollments based on fuzzy time series. Fuzzy Sets Syst. 1996; 81(3): 311-9. https://doi.org/10.1016/0165-0114(95)00220-0
Yu HK. Weighted fuzzy time series models for taiex forecasting. Physica A. 2005; 349(3-4): 609-24. https://doi.org/10.1016/j.physa.2004.11.006
Huarng K, Yu TH. Ratio-based lengths of intervals to improve fuzzy time series forecasting. IEEE Trans Cybern. 2006; 36(2): 328-40. https://doi.org/10.1109/TSMCB.2005.857093
Cheng C, Chen T, Teoh HJ, Chiang CH. Fuzzy time-series based on adaptive expectation model for taiex forecasting. Expert Syst Appl. 2008; 34(2): 1126-32. https://doi.org/10.1016/j.eswa.2006.12.021
Yolcu U, Egrioglu E, Uslu VR, Basaran MA, Aladag CH. A new approach for determining the length of intervals for fuzzy time series. Appl Soft Comput. 2009; 9(2): 647-51. https://doi.org/10.1016/j.asoc.2008.09.002
Tsaur RC. A fuzzy time series-markov chain model with an application to forecast the exchange rate between the taiwan and us dollar. Int J Innov Comput Inf Control. 2012; 8(7): 4931-42.
Guney H, Bakir MA, Aladag CH. A novel stochastic seasonal fuzzy time series forecasting model. Int J Fuzzy Syst. 2018; 20: 729-40. https://doi.org/10.1007/s40815-017-0385-z
Song Q. Seasonal forecasting in fuzzy time series. Fuzzy Sets Syst. 1999; 107(2): 235-6. https://doi.org/10.1016/S0165-0114(98)00266-8
Egrioglu E, Aladag CH, Yolcu U, Basaran MA, Uslu VR. A new hybrid approach based on sarima and partial high order bivariate fuzzy time series forecasting model. Expert Syst Appl. 2009; 36(4): 7424-34. https://doi.org/10.1016/j.eswa.2008.09.040
Suhartono, Lee MH, Javedani H. A weighted fuzzy integrated time series for forecasting tourist arrivals. In: Abd Manaf A, Zeki A, Zamani M, Chuprat S, El-Qawasmeh E, Eds. Informatics Engineering and Information Science. ICIEIS 2011. Communications in Computer and Information Science, vol 252. Berlin, Heidelberg: Springer; 2011, pp. 206-17. https://doi.org/10.1007/978-3-642-25453-6_19
Tseng FM, Tzeng GH. A fuzzy seasonal arima model for forecasting. Fuzzy Sets Syst. 2002; 126(3): 367-76. https://doi.org/10.1016/S0165-0114(01)00047-1
Box G. Box and jenkins: time series analysis, forecasting and control. In A Very British Affair: Six Britons and the Development of Time Series Analysis During the 20th Century. London: Palgrave Macmillan, Springer; 2013, pp. 161-215. https://doi.org/10.1057/9781137291264_6
Hao-Tien Liu and Mao-Len Wei. An improved fuzzy forecasting method for seasonal time series. Expert Syst Appl. 2010; 37(9): 6310-8. https://doi.org/10.1016/j.eswa.2010.02.090
Tsymbal A. The problem of concept drift: definitions and related work. Computer Science Department, Trinity College Dublin, 2004; 106(2): 58.
Ditzler G, Roveri M, Alippi C, Polikar R. Learning in nonstationary environments: A survey. IEEE Comput Intell Mag. 2015; 10(4): 12-25. https://doi.org/10.1109/MCI.2015.2471196
Song Q, Chissom BS. Forecasting enrollments with fuzzy time series-part ii. Fuzzy Sets Syst. 1994; 62(1): 1-8. https://doi.org/10.1016/0165-0114(94)90067-1
AJ Saleena and C Jessy John. New hybrid model based on triple exponential smoothing and fuzzy time series for forecasting seasonal time series. In: Deo N, Gupta V, Acu A, Agrawal P, Eds. Mathematical analysis ii: optimisation, differential equations and graph theory. ICRAPAM 2018. Springer Proceedings in Mathematics & Statistics, vol 307. Singapore: Springer; 2018, pp. 179-90. https://doi.org/10.1007/978-981-15-1157-8_16
Singh SR. A simple time variant method for fuzzy time series forecasting. Cybern Syst. 2007; 38(3): 305-21. https://doi.org/10.1080/01969720601187354
Wei LY. A hybrid anfis model based on empirical mode decomposition for stock time series forecasting. Appl Soft Comput. 2016; 42: 368-76. https://doi.org/10.1016/j.asoc.2016.01.027
Jilani TA, Burney SMA. A refined fuzzy time series model for stock market forecasting. Physica A. 2008; 387(12): 2857-62. https://doi.org/10.1016/j.physa.2008.01.099
Singh P. A novel hybrid time series forecasting model based on neutrosophic-pso approach. Int J Mach Learn Cybern. 2020; 11(8): 1643-58. https://doi.org/10.1007/s13042-020-01064-z
Singh P, Dhiman G. A hybrid fuzzy time series forecasting model based on granular computing and bio-inspired optimization approaches. J Comput Sci. 2018; 27: 370-85. https://doi.org/10.1016/j.jocs.2018.05.008
Singh P. A brief review of modeling approaches based on fuzzy time series. Int J Mach Learn Cybern. 2017; 8(2): 397-420. https://doi.org/10.1007/s13042-015-0332-y
Bose M, Mali K. Designing fuzzy time series forecasting models: A survey. Int J Approx Reason. 2019; 111: 78-99. https://doi.org/10.1016/j.ijar.2019.05.002
Vovan T. An improved fuzzy time series forecasting model using variations of data. Fuzzy Optim Decis Making. 2019; 18: 151-73. https://doi.org/10.1007/s10700-018-9290-7
Alves MA, Silva PCL, Junior CAS, Vieira GL, Guimarães FG, Sadaei HJ. An extension of nonstationary fuzzy sets to heteroskedastic fuzzy time series. In ESANN 2018 proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges (Belgium), 25-27 April 2018.
Silva PCL, Junior CAS, Alves MA, Silva R, Cohen MW, Guimarães FG. Forecasting in non-stationary environments with fuzzy time series. Appl Soft Comput. 2020; 97: 106825. https://doi.org/10.1016/j.asoc.2020.106825
Arnold M, Milner XHR, Witte H, Bauer R, Braun C. Adaptive ar modeling of nonstationary time series by means of kalman filtering. IEEE Trans Biomed Eng. 1998; 45(5): 553-62. https://doi.org/10.1109/10.668741
Kitagawa G, Gersch W. A smoothness priors time-varying ar coefficient modeling of nonsta- tionary covariance time series. IEEE Trans Automat Contr. 1985; 30(1): 48-56. https://doi.org/10.1109/TAC.1985.1103788
Bitencourt HV, Souza LAF, Santos MC, Silva R, Lima PC, Guimarães FG, et al. Combining embeddings and fuzzy time series for high-dimensional time series forecasting in internet of energy applications. Energy. 2023; 271: 127072. https://doi.org/10.1016/j.energy.2023.127072
Bitencourt HV, Orang O, Souza LAF, Silva PCL, Guimarães FG. An embedding-based non-stationary fuzzy time series method for multiple output high-dimensional multivariate time series forecasting in iot applications. Neural Comput Appl. 2023; 35(13): 9407-20. https://doi.org/10.1007/s00521-022-08120-5
Pattanayak RM, Behera HS, Panigrahi S. A novel probabilistic intuitionistic fuzzy set based model for high order fuzzy time series forecasting. Eng Appl Artif Intell. 2021; 99: 104136. https://doi.org/10.1016/j.engappai.2020.104136
Dixit A, Jain S. Intuitionistic fuzzy time series forecasting method for non-stationary time series data with suitable number of clusters and different window size for fuzzy rule generation. Inf Sci. 2023; 623: 132-45. https://doi.org/10.1016/j.ins.2022.12.015
Yolcu OC, Yolcu U. A novel intuitionistic fuzzy time series prediction model with cascaded structure for financial time series. Expert Syst Appl. 2023; 215: 119336. https://doi.org/10.1016/j.eswa.2022.119336
Kocak C, Egrioglu E, Bas E. A new explainable robust high-order intuitionistic fuzzy time-series method. Soft Comput. 2023; 27(3):1783-96. https://doi.org/10.1007/s00500-021-06079-4
Radha Mohan Pattanayak, Himansu Sekhar Behera, and Sibarama Panigrahi. A novel high order hesitant fuzzy time series forecasting by using mean aggregated membership value with support vector machine. Inf Sci. 2023; 626: 494-523. https://doi.org/10.1016/j.ins.2023.01.075
Pant M, Kumar S. Fuzzy time series forecasting based on hesitant fuzzy sets, particle swarm optimization and support vector machine-based hybrid method. Granular Comput. 2022; 7(4): 861-79. https://doi.org/10.1007/s41066-021-00300-3
Farid M. The subgradient extragradient method for solving mixed equilibrium problems and fixed point problems in hilbert spaces. J Appl Numer Optim. 2019; 1(3): 335-45. https://doi.org/10.23952/jano.1.2019.3.10
Farid M, Kazmi KR. A new mapping for finding a common solution of split generalized equilibrium problem, variational inequality problem and fixed point problem. Korean J Math. 2019; 27(2): 297-327.
Djafari-Rouhani B, Kazmi KR, Farid M. Common solutions to some systems of variational inequalities and fixed point problems. Fixed Point Theory. 2017; 18(1): 167-90. https://doi.org/10.24193/fpt-ro.2017.1.14
Alansari M, Farid M, Ali R. An iterative scheme for split monotone variational inclusion, variational inequality and fixed point problems. Adv Differ Equ. 2020: 485: 1-21. https://doi.org/10.1186/s13662-020-02942-0
Zadeh LA. Outline of a new approach to the analysis of complex systems and decision processes IEEE Trans Cybern. 1973; (1): 28-44. https://doi.org/10.1109/TSMC.1973.5408575
Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning-i. Infr Sci. 1975; 8(3): 199-249. https://doi.org/10.1016/0020-0255(75)90036-5
Song Q, Chissom BS. Fuzzy time series and its models. Fuzzy Sets Syst., 1993; 54(3): 269-77. https://doi.org/10.1016/0165-0114(93)90372-O
Song Q, Chissom BS. Forecasting enrollments with fuzzy time series-part i. Fuzzy Sets Syst. 1993; 54(1): 1-9. https://doi.org/10.1016/0165-0114(93)90355-L
Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997; 9(8): 1735-80. https://doi.org/10.1162/neco.1997.9.8.1735
MacQueen J. Some Methods for Classification and Analysis of Multivariate Observations. Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, vol. 1, Oakland, CA, USA: 1967, pp. 281-97.
Chen SM, Tanuwijaya K. Fuzzy forecasting based on high-order fuzzy logical relationships and automatic clustering techniques. Expert Syst Appl. 2011; 38(12): 15425-37. https://doi.org/10.1016/j.eswa.2011.06.019
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright (c) 2024 Abdul J. Saleena, John C. Jessy , M.C. Lineesh