FİNANSAL VERİSETLERİ İÇİN BOZKURT OPTİMİZASYON TEMELLİ GERİ BESLEMELİ BULANIK ÇIKARIM FONKSİYONLARI

Year 2020, Volume: 15 Issue: 54, 350 - 366, 30.07.2020

Nihat Tak

Abstract

Zaman serisi
modelleri, tıp, mühendislik, işletme, ekonomi ve finans gibi birçok alanda,
önceki dönemlerden gözlem değerleri yardımıyla tahminler yapmak amacıyla yaygın
olarak kullanılmaktadır. Bu nedenle, özellikle alternatif/olasılık dışı
yöntemler kullanılarak, zaman serisi tahmin performanslarını geliştirmek için
birçok çaba vardır. Bu çalışmada, zaman serisi veri kümesindeki doğrusal
olmayan yapının üstesinden gelebilmek için, Bozkurt optimizasyon (GWO) temelli
Otoregresif hareketli ortalama (ARMA) modeli ile tip-1 bulanık fonksiyonların (T1FFs)
birleştirilmesiyle yeni bir tahmin yaklaşımı önerilmiştir. GWO'nun, arama
boyunca keşif ve uygun stabiliteye hızlı ulaşması, daha az depolama
gereksinimleri ve hızlı yakınsama gibi diğer yöntemler üzerindeki üstünlükleri
göz önüne alındığında, kare hatalarının toplamını en aza indirgemek için
geribeslemeli T1FFs yönteminin katsayılarının tahmini GWO ile elde edilmesi
uygun görüşmüştür. Beş farklı gerçek veri kümesinde önerilen yöntemin ve mevcut
birkaç tahmin yönteminin karşılaştırılması gerçekleştirilmiştir. Sonuçlar,
önerilen yöntemin, ortalama mutlak yüzde hataları ve kök ortalama kare hataları
ile birlikte daha iyi çalışma süresi açısından çoğu zaman daha iyi tahminler
ürettiğini göstermektedir

Keywords

Tip-1 Bulanık Fonksiyonlar, Boz-kurt Optimizasyonu, Öngörü, Otoregresif Hareketli Ortalamlar

GREY WOLF OPTIMIZER BASED RECURRENT FUZZY REGRESSION FUNCTIONS FOR FINANCIAL DATASETS

Abstract

Time
series models are used extensively in many fields, such as medicine,
engineering, business, economics and finance, with the aim of making forecasts
through the help of observation values from previous periods. Therefore, there
are many efforts to improve time series forecasting performances in the recent
literature, mainly using alternative/non-probabilistic methods. In the present
study, a novel forecasting approach has been proposed by combining the type-1
fuzzy functions (T1FF) with the Autoregressive moving average (ARMA) model
based on grey wolf optimizer (GWO) in order to be able to overcome the
nonlinear structure in time series dataset. Considering the superiorities of
GWO over other methods, such as less storage requirements and rapid convergence
by striking the proper stability between the exploration and exploitation
throughout the search, estimation of the coefficients of the R-T1FFs method
obtained through GWO to minimize the sum of squared errors (SSE). Comparison of
the proposed method and several existing forecasting methods has been performed
on five real world time series datasets. The results indicate that the proposed
method produces better forecasts most of the time in the terms of mean absolute
percentage errors and root mean square errors along with the better running
time.

Keywords

Type-1 Fuzzy Functions, Grey Wolf Optimizer, Autoregressive Moving Average, Forecasting

References

• Aladag, C.H., Turksen, I.B., Dalar, A.Z., Egrioglu, E. & Yolcu, U. (2014). Application of type-1 fuzzy functions approach for time series forecasting, Turkish Journal of Fuzzy Systems, 5(1), 1-9.
• Aladag, C.H., Yolcu, U., Egrioglu. E. & Dalar, A.Z. (2012). A new time invariant fuzzy time series method based on particle swarm optimization, Applied Soft Computing, 12(10), 3291-3299.
• Aladag, C.H., Yolcu, U. & Egrioglu, E. (2015). A new multiplicative seasonal neural network model based on particle swarm optimization, Neural Processing Letters, 37(3), 251-262.
• Bas, E., Egrioglu, E., Yolcu, U. & Aladag, C.H. (2015). Fuzzy time series network used to forecast linear and nonlinear time series, Applied Intelligence, 43(2), 343-355.
• Beyhan, S. & Alci, M. (2010). Fuzzy functions based arx model and new fuzzy basis function models for nonlinear system identification, Applied Soft Computing, 10(2), 439-444.
• ISEX. (2015). Istanbul stock exchange index dataset. http://www.borsaistanbul.com/veriler/gecmise-donukveri- satisi. (Accessed 5 November 2015).
• Celikyilmaz, A. & Turksen, B. (2009). Modeling Uncertainty with Fuzzy Logic: With Recent Theory and Applications. Berlin: Springer.
• Chang, B.R. (2008). Resolving the forecasting problems of overshoot and volatility clustering using ANFIS coupling nonlinear heteroscedasticity with quantum tuning, Fuzzy Sets and Systems, 159(23), 3183- 3200.
• Chau, K.W. (2006). Particle swarm optimization-training algorithm for ANNs in stage prediction of shing mun river, Journal of Hydrology, 329(3, 4), 363-367.
• Chen, S. M. (1996). Forecasting enrollments based on fuzzy time series. Fuzzy sets and systems, 81(3), 311-319.
• Chen, S. M., & Chang, Y. C. (2010). Multi-variable fuzzy forecasting based on fuzzy clustering and fuzzy rule interpolation techniques. Information sciences, 180(24), 4772-4783.
• Chen, S. M., & Chen, C. D. (2011). TAIEX forecasting based on fuzzy time series and fuzzy variation groups. IEEE Transactions on Fuzzy Systems, 19(1), 1-12.
• Chen, S. M., Chu, H. P., & Sheu, T. W. (2012). TAIEX forecasting using fuzzy time series and automatically generated weights of multiple factors. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans, 42(6), 1485-1495.
• Chen, D. & Zhang, J. (2005). Time series prediction based on ensemble ANFIS, Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, 3552-3556.
• Eberhart, R. & Kennedy, J. (1995). A new optimizer using particle swarm theory, Micro Machine and Human Science Proceedings of the Sixth International Symposium on. IEEE (MHS’95).
• Egrioglu, E., Aladag, C.H., Yolcu, U. & Bas, E. (2014). A new adaptive network based fuzzy inference system for time series forecasting, Aloy Journal of Soft Computing and Applications, 2(1), 25-32.
• Faris, H., Aljarah, I., Al-Betar, M.A. & Mirjalili, S. (2018). Grey wolf optimizer: A review of recent variants and applications, Neural Computing and Applications, 30(2), 413-435.
• Hong, W.C. (2009). Chaotic particle swarm optimization algorithm in a support vector regression electric load forecasting model, Energy Conversion and Management, 50(1), 105-117.
• Hong, W.C. (2010). Application of chaotic ant swarm optimization in electric load forecasting, Energy Policy, 38(10), 5830-5839.
• Huang, C.M., Huang, C.J. & Wang, M.L. (2005). A particle swarm optimization to identifying the ARMAX model for short-term load forecasting, IEEE Transactions on Power Systems, 20(2), 1126-1133.
• Janacek, G. & Janacek, G. J. (2001). Practical Time Series. London: Arnold.
• Jang, J.S.R. (1993). ANFIS: Adaptive-network-based fuzzy inference system, IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665-685.
• Karaboga, D. & Basturk, B. (2007). A powerful and efficient algorithm for numerical function optimization: Artificial bee colony (ABC) algorithm, Journal of Global Optimization, 39(3), 459-471.
• Kumaran, J. & Ravi, G. (2014). Long-term forecasting of electrical energy using ANN and HSA. International Review on Modelling and Simulations (IREMOS), 7(3), 489-496.
• Liao, G. & Tsao, T. (2006). Application of a fuzzy neural network combined with a chaos genetic algorithm and simulated annealing to short-term load forecasting, IEEE Transactions on Evolutionary Computation, 10(3), 330-340.
• Mirjalili, S.M.S. & Lewis, A. (2014) Grey wolf optimizer, Advances in Engineering Software, 69,496-517.
• Mustaffa, Z., Sulaiman, M.H. & Kahar, M.N.M. (2015, August). LS-SVM hyper-parameters optimization based on GWO algorithm for time series forecasting. In Software Engineering and Computer Systems (ICSECS), 2015 4th International Conference on (pp. 183-188). IEEE.
• Niu, D., Wang, Y. & Wu, D.D. (2010). Power load forecasting using support vector machine and ant colony optimization, Expert Systems with Applications, 37(3), 2531-2539.
• Niu, M., Wang, Y., Sun, S. & Li, Y. (2016). A novel hybrid decomposition-and-ensemble model based on CEEMD and GWO for short-term PM 2.5 concentration forecasting, Atmospheric Environment, 134,168-180.
• Pradhan, M., Roy, P.K. & Pal, T. (2016). Grey wolf optimization applied to economic load dispatch problems, International Journal of Electrical Power and Energy Systems, 83, 325-334.
• Saad, A.E.H., Dong, Z. & Karimi, M. (2017). A comparative study on recently-introduced nature-based global optimization methods in complex mechanical system design, Algorithms, 10(4), 120.
• Wei, L. (2016). A hybrid anfis model based on empirical model decomposition for stock time series forecasting, Applied Soft Computing, 42, 368-376.
• Mamdani, E.H. & Assilian, S. (1975). An experiment in linguistic synthesis with a fuzzy logic controller, International Journal of Man-Machine Studies, 7(1), 1-13.
• TAIEX. (2015). Taiwan stock exchange index dataset. http://www.taiwanindex.com.tw/index/history/t00. Accessed 17 October 2015.
• Tak , N., Evren, A.A., Tez, M. & Egrioglu, E. (2018). Recurrent type-1 fuzzy functions approach for time series forecasting, Applied Intelligence, 48(1), 68-77, doi: 10.1007/s10489.017.0962-8.
• Tak , N. (2018). Meta fuzzy functions: Application of recurrent type-1 fuzzy functions, Applied Soft Computing, 73, 1-13.
• Tak, N. (2020). Type-1 recurrent intuitionistic fuzzy functions for forecasting, Expert Systems with Applications, 140, 112913.
• Takagi, T. & Sugeno, M. (1985). Fuzzy identification of systems and its applications to modeling and control, IEEE Transactions on Systems, Man, and Cybernetics, 15(1), 116-132.
• Sarica, B., Egrioglu, E. & Asikgil, B. (2016). A new hybrid method for time series forecasting, Neural Computing and Applications, 29(3), 749-760.
• Yusof, Y. & Mustaffa, Z. (2015, March). Time series forecasting of energy commodity using grey wolf optimizer, In Proceedings of the International Multi Conference of Engineers and Computer Scientists (IMECS’15) (Vol. 1).
• Zou, Z.D., Sun, Y.M. & Zhang, Z.S. (2005). Short-term load forecasting based on recurrent neural network using ant colony optimization algorithm, Power System Technology, 3, 59-63.