Research Article
BibTex RIS Cite

LSTM Mimarisi Kullanarak USD/TRY Fiyat Tahmini

Year 2020, Ejosat Special Issue 2020 (ARACONF), 452 - 456, 01.04.2020
https://doi.org/10.31590/ejosat.araconf59

Abstract

Son zamanlarda, derin öğrenme yaklaşımlarının hızlı bir şekilde gelişmesi bu konuya olan ilgiyi arttırmış ve birçok alanda başarılı bir şekilde uygulanmaya başlanmıştr. Bu alanlardan birisi de finansal zaman verileridir. Finansal varlıkların fiyatını tahmin etmek, doğru tahminlerle yatırım karar verme riskini azaltabileceğinden önemlidir. LSTM (Uzun kısa süreli bellek), zaman serilerindeki önemli aralık ve uzun gecikme olaylarını işleyip tahmin etmek için uygun ve sıralı verilerde kullanılan yeni bir algoritmadır. Değerlendirmeler 1/1/2000 - 12/31/2017 tarihleri arasında USD/TRY paritesi veri seti kullanılarak gerçekleştirilmiştir.Yapılan çalışmalar sonucunda LSTM yaklaşımının başarılı, gerçek değerlere daha yakın bir tahmin yaptığı görülmüştür. Bunun nedeni LSTM mimarisinin dahili bir belleğe sahip olup girişini hatırlayabilmesidir. Bu makale de LSTM mimarisinin zamansal özelliklere dayanmasından dolayı zamansal verilerin (stok verileri, finansal veriler vb) tahmin sürecinde başarılı bir şekilde uygulanabilir olduğu gözlenmiştir.

References

  • William W. S. Wei, (2006). Time Series Analysis: Univariate and Multivariate Methods, Pearson Addison Wesley
  • Cavalcante, R.C., Brasileiroi, R C., Souza, V.L, Nobrega, J.P., Oliveira, A.L.,(2016).Computational intelligence and financial markets: a survey and future directions, Expert Systems With Applications,55,194-211.
  • Maimonand, O., Rokach, L.,(2005). Data Mining An dKnowledge Discovery Handbook. New York, NY, USA: Springer,doi: 10.1007/b107408.
  • Hiransha, M., Gopalakrishnan, E.A., Vijay Krishna Menonab, K.P. Soman, (2018). NSE stock market prediction using deep-learning models, Procedia Computer Science, 132,1351–1362
  • Fischer, T., Krauss, C., (2018). Deep learning with long short-term memory networks for financial market predictions, European Journal of Operational Research, 270, 654–669.
  • Cheng, L.C., Huang, Y.H., Wu, M.E., (2018). Applied attention-based LSTM neural networks in stock prediction, IEEE International Conference on Big Data (Big Data), Seattle, WA, USA

USD / TRY Price Prediction Using LSTM Architecture

Year 2020, Ejosat Special Issue 2020 (ARACONF), 452 - 456, 01.04.2020
https://doi.org/10.31590/ejosat.araconf59

Abstract

Recently, the rapid development of deep learning approaches has increased the interest in this subject and has started to be applied successfully in many areas. One of these areas is financial time data. Prediction a financial asset's price is important as one can lower the risk of investment decision- making with accurate prediction. LSTM (Term Memory Long-Short) is suitable for processing and predicting the important events of interval and long delay in time series and a new algorithm used in sequential data. The evaluations was conducted using between 1/1/2000 - 12/31/2017 using USD / TRY parity dataset. As a result of the studies, it was seen that the LSTM approach made successful and closer predict to the real values. This is because the LSTM architecture has an internal memory and can remember its input. In this article, it has been observed that temporal data (stock data, financial data etc.) can be applied successfully in the prediction process since LSTM architecture is based on temporal properties.

References

  • William W. S. Wei, (2006). Time Series Analysis: Univariate and Multivariate Methods, Pearson Addison Wesley
  • Cavalcante, R.C., Brasileiroi, R C., Souza, V.L, Nobrega, J.P., Oliveira, A.L.,(2016).Computational intelligence and financial markets: a survey and future directions, Expert Systems With Applications,55,194-211.
  • Maimonand, O., Rokach, L.,(2005). Data Mining An dKnowledge Discovery Handbook. New York, NY, USA: Springer,doi: 10.1007/b107408.
  • Hiransha, M., Gopalakrishnan, E.A., Vijay Krishna Menonab, K.P. Soman, (2018). NSE stock market prediction using deep-learning models, Procedia Computer Science, 132,1351–1362
  • Fischer, T., Krauss, C., (2018). Deep learning with long short-term memory networks for financial market predictions, European Journal of Operational Research, 270, 654–669.
  • Cheng, L.C., Huang, Y.H., Wu, M.E., (2018). Applied attention-based LSTM neural networks in stock prediction, IEEE International Conference on Big Data (Big Data), Seattle, WA, USA
There are 6 citations in total.

Details

Primary Language Turkish
Subjects Engineering
Journal Section Articles
Authors

Özlem Alpay This is me 0000-0002-5432-4102

Publication Date April 1, 2020
Published in Issue Year 2020 Ejosat Special Issue 2020 (ARACONF)

Cite

APA Alpay, Ö. (2020). LSTM Mimarisi Kullanarak USD/TRY Fiyat Tahmini. Avrupa Bilim Ve Teknoloji Dergisi452-456. https://doi.org/10.31590/ejosat.araconf59