EURO/TL KURU TAHMİNİNDE İSTATİSTİK VE YAPAY SİNİR AĞLARI KULLANIMI

Year 2018, Volume: 7 Issue: 1, 414 - 417, 01.09.2018

Oktay Tas , Emir Yakak Umut Ugurlu

https://doi.org/10.17261/Pressacademia.2018.926

Abstract

Amaç- Bu çalışmanın amacı Euro/Türk lirası kurunun hareketinin istatistik ve yapay sinir ağları yöntemleri ile tahmin edilmesidir.

Yöntem- Çalışmada iki farklı tahmin yöntemi ile Euro/Türk lirası kuru tahmini yapılmıştır. Girdi olarak her iki modelde de son 10 yılın Euro/Türk lirası günlük kuru kullanılmış ve son 1 yılın günlük dolar kuru tahmin edilmiştir.

Bulgular- Yapay sinir ağları yöntemi ile bulunan ortalama mutlak hatalar istatistik yöntemi ile bulunanların yaklaşık %2’si kadar daha azdır. Tahminler 365 günün her biri için “rolling window” yöntemi kullanılarak yapıldığından, elde edilen sonuçların “robust” olduğu söylenebilir.

Sonuç- Araştırmada kullanılan her iki modelin de belirli bir başarı ile Dolar kurunu tahmin tahmin edebildikleri ancak Yapay Sinir Ağları modelinin, istatistik modeline kıyasla daha başarılı sonuçlar verdiği gözlemlenmiştir. Bundan sonraki çalışmalarda dışsal değişkenlerin de modele eklenmesi ile tahmin performansının arttırılabilmesi mümkün olabilir.

Keywords

Döviz kuru, yapay sinir ağlari, regresyon, zaman serisi tahmini, avro

USING ARTIFICIAL NEURAL NETWORK AND A STATISTICAL METHOD FOR THE ESTIMATION OF EURO/TURKISH LIRA EXCHANGE RATE

Abstract

Purpose- The aim of this study is to estimate the movement of the Euro/Turkish lira currency with a statistical method and artificial neural networks methods and to compare the performance of these two methods.

Methodology- In the study, two different forecasting methods were used to estimate the dollar exchange rate. In both models, the Euro/Turkish lira daily rate for the last 10 years was used and the daily dollar rate for the last 1 year was estimated.

Findings- The mean absolute errors found by artificial neural networks method are about 2% less than those found by the statistical method. Since estimates are made using the "rolling window" method for each of the 365 days, it can be said that the results obtained are "robust".

Conclusion- It has been observed that the Artificial Neural Networks model yields more successful results than the statistical model, although both models used in the research can forecast the dollar exchange rate with a certain success. In future studies it may be possible to increase the estimation performance by adding the exogenous variables to the model.

Keywords

Exchange rate, artificial neural networks, regression, time series forecast, euro

References

• Dash, M., Liu, H. (1997). Feature selection for classification. Intelligent data analysis, 1(3), 131-156.
• Fausett, L. V. (1994). Fundamentals of neural networks: architectures, algorithms, and applications (Vol. 3). Englewood Cliffs: Prentice-Hall.
• Gallant, S. I. (1993). Neural network learning and expert systems. MIT press.
• Hann, T. H., Steurer, E. (1996). Much ado about nothing? Exchange rate forecasting: neural networks vs. linear models using monthly and weekly data. Neurocomputing, 10(4), 323-339.
• Kamruzzaman, J., Sarker, R. A. (2003). Forecasting of currency exchange rates using ANN: a case study. In Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on (Vol. 1, pp. 793-797). IEEE.
• Kuan, C. M., Liu, T. (1995). Forecasting exchange rates using feedforward and recurrent neural networks. Journal of applied econometrics, 10(4), 347-364.
• Weigend, A. S., Huberman, B. A., Rumelhart, D. E. (1992). Predicting sunspots and exchange rates with connectionist networks. PRE-33772.
• Wu, B. (1995). Model-free forecasting for nonlinear time series (with application to exchange rates). Computational Statistics & Data Analysis, 19(4), 433-459.
• Zhang, G., Hu, M. Y. (1998). Neural network forecasting of the British pound/US dollar exchange rate. Omega, 26(4), 495-506.
• Zhang, G., Patuwo, B. E., Hu, M. Y. (1998). Forecasting with artificial neural networks: the state of the art. International journal of forecasting, 14(1), 35-62.