ALTIN FİYATI GÜNLÜK GETİRİLERİNİN YAPAY SİNİR AĞLARI ALGORİTMASI VE MARKOV ZİNCİRLERİ MODELLERİ İLE TAHMİNİ

Year 2018, 18. EYI Special Issue, 681 - 694, 20.01.2018

Salih Çam , Süleyman Bilgin Kılıç

https://doi.org/10.18092/ulikidince.347048

Cited By: 6

Abstract

Son
on yılda altın fiyatları ile ilgili birçok çalışma yapılmıştır. Literatürdeki ilgili
çalışmalar yapay sinir ağları algoritması ve Markov zincirleri modellerinin
altın piyasası gelecek tahmininde oldukça iyi sonuçlar ürettiklerini
göstermektedir. Bu çalışmada yapay sinir ağları algoritması ve Markov
zincirleri modellerinin güçlü yönleri birlikte kullanılarak altın fiyatlarının günlük
getirileri tahmin edilmiştir. Tahmin süreci, iki aşamada gerçekleştirilmiştir.
İlk aşamada altın fiyatlarının günlük getirileri, en iyi tahmin performansına
sahip yapay sinir ağları algoritması ile tahmin edilmiştir. İkinci aşamada ise yapay
sinir ağlarından elde edilen tahmini altın getirileri üzerinden yüksek
dereceden Markov zincirleri geçiş olasılıkları matrisleri hesaplanmıştır. Tahmin,
üçüncü dereceden ve dördüncü dereceden Markov zincirleri modelleri ile yapay
sinir ağları algoritması birlikte kullanılarak yapılmıştır ve uygulanan tahmin
yöntemi sonucunda altın fiyatları getirilerinin gelecek tahmininde %70’i bulan başarı
sağlanmıştır.

Keywords

Yapay Sinir Ağları, Markov Zincirleri Modelleri, Altın Fiyatları

PREDICTION OF DAILY GOLD PRICE RETURNS WITH ARTIFICIAL NEURAL NETWORKS AND MARKOV CHAINS MODELS

Abstract

During the last
decade, there have been many studies on the prediction
of gold prices. Artificial neural networks algorithm and Markov chains models
perform fairly well in the prediction of gold prices. In this study, the daily
returns of gold prices were estimated by using the powerful aspects of the
artificial neural network algorithm and the Markov chains models. The
prediction was made in two stages. In the first stage, the daily return series
of gold prices were estimated by the artificial neural network algorithm which has
the best prediction performance. In the second stage, the transition
probabilities matrixes of high order Markov chains models were calculated from
the estimated values that obtained from the artificial neural networks.  The third order and the fourth order Markov
chains models and the artificial neural networks algorithm were used jointly
for prediction. In consequence of the applied methodology, about 70% success of
prediction was achieved in the future prediction of the gold prices’ returns.

Keywords

Artificial Neural Network, Markov Chains Models, Gold Price

References

• Altan, Ş. (2008). Döviz Kuru Öngörü Performansı İçin Alternatif Bir Yaklaşım:Yapay Sinir Ağı. İktisadi ve İdari Bilimler Fakültesi Dergisi, 10(2), 1-20.
• Aydin, A. D., & Cavdar, S. C. (2015). Comparison of prediction performances of artificial neural network (ANN) and vector autoregressive (VAR) Models by using the macroeconomic variables of gold prices, Borsa Istanbul (BIST) 100 index and US Dollar-Turkish Lira (USD/TRY) exchange rates. Procedia Economics and Finance, 30, 3-14.
• Berchtold, A., & Raftery, A. E. (2002). The mixture transition distribution model for high-order Markov chains and non-Gaussian time series.Statistical Science, 328-356.
• Can, T., & Öz, E. (2009). Saklı Markov modelleri kullanılarak Türkiye'de dolar kurundaki değişimintahmin edilmesi. Istanbul University Journal of the School of Business Administration, 38(1).
• Godarzi, A. A., Amiri, R. M., Talaei, A., & Jamasb, T. (2014). Predicting oil price movements: A dynamic Artificial Neural Network approach. Energy Policy, 68, 371-382.
• Hamilton, J. D. (1989). A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica: Journal of the Econometric Society, 357-384.
• Hamilton, J. D. (1994). Time series analysis (Vol. 2). Princeton: Princeton university press.
• Kamruzzaman, J., & Sarker, R. A. (2003, December). 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.
• Kanas, A. (2003). Non‐linear forecasts of stock returns. Journal of Forecasting, 22(4), 299-315
• Kanas, A., & Genius, M. (2005). Regime (non) stationarity in the US/UK real exchange rate. Economics Letters, 87(3), 407-413.
• Kaynar, O., & Taştan, S. (2009). Zaman Serisianalizinde Mlp Yapay Sinir Ağları Ve Arıma Modelinin Karşılaştırılması. Erciyes Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, (33), 161-172.
• Khashei, M., Hejazi, S. R., & Bijari, M. (2008). A new hybrid artificial neural networks and fuzzy regression model for time series forecasting. Fuzzy sets and systems, 159(7), 769-786.
• Kılıç, S. B. (2013). Integrating Artificial Neural Network Models By Markov Chain Process: Forecasting The Movement Direction Of Turkish Lira Us Dollar Exchange Rate Returns. Çukurova Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 22(2).
• Kim, C. J., Piger, J., & Startz, R. (2008). Estimation of Markov regime-switching regression models with endogenous switching. Journal of Econometrics, 143(2), 263-273.
• Kristjanpoller, W., & Minutolo, M. C. (2016). Forecasting volatility of oil price using an artificial neural network-GARCH model. Expert Systems with Applications, 65, 233-241.
• Leung, M. T., Chen, A. S., & Daouk, H. (2000). Forecasting exchange rates using general regression neural networks. Computers & Operations Research, 27(11), 1093-1110.
• Majhi, R., Panda, G., & Sahoo, G. (2009). Efficient prediction of exchange rates with low complexity artificial neural network models. Expert Systems with Applications, 36(1), 181-189.
• Malkiel, B. G., & Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. The journal of Finance, 25(2), 383-417. Mamipour, S., & Vaezi Jezeie, F. (2015). Non-Linearities in the relation between oil price, gold price and stock market returns in Iran: a multivariate regime-switching approach.
• Marsh, I. W. (2000). High‐frequency Markov switching models in the foreign exchange market. Journal of Forecasting, 19(2), 123-134.
• McQueen, G., & Thorley, S. (1991). Are stock returns predictable? A test using Markov chains. The Journal of Finance, 46(1), 239-263.
• Mills, T. C., & Jordanov, J. V. (2003). The size effect and the random walk hypothesis: Evidence from the London Stock Exchange using Markov chains. Applied Financial Economics, 13(11), 807-815.
• Nag, A. K., & Mitra, A. (2002). Forecasting daily foreign exchange rates using genetically optimized neural networks. Journal of Forecasting, 21(7), 501-511.
• Öz, E. (2009). İstanbul Menkul Kıymetler Borsası Üzerine Saklı Markov Modeli İle Bir Tahminleme. Ekonomik Yaklasim, 20(72), 59-85.
• Özkan, F. (2012). Döviz Kuru Tahmininde Parasal Model ve Yapay Sinir Aglari Karsilastirmasi. Business and Economics Research Journal, 3
• Öztürk, A. (2014). Yöneylem araştırması. Ekin Kitabevi Yayınları, BURSA
• Paksoy, S., & Kilic, S. B. (2015). Forecasting the Direction of BIST 100 Returns with Artificial Neural Network Models. International Journal of Latest Trends in Finance and Economic Sciences, 4(3), 7.
• Panda, C., & Narasimhan, V. (2007). Forecasting exchange rate better with artificial neural network. Journal of Policy Modeling, 29(2), 227-236.
• Raftery, A. E. (1985). A model for high-order Markov chains. Journal of the Royal Statistical Society. Series B (Methodological), 528-539.
• Rosenblatt, M., & Slepian, D. (1962). N th Order Markov Chains with Every N Variables Independent. Journal of the Society for Industrial and Applied Mathematics, 10(3), 537-549.