Derin Öğrenme Modellerinin Gümüş Fiyat Tahmininde Karşılaştırmalı Analizi: CNN, LSTM, GRU ve Hibrit Yaklaşım

Year 2024, Volume: 24 Issue: 1, 1 - 13, 22.05.2024

Yunus Emre Gür

https://doi.org/10.25294/auiibfd.1404173

Abstract

Bu çalışmada, gümüş fiyatlarını tahmin etmek amacıyla farklı derin öğrenme algoritmalarının performansını değerlendirilmiştir. Tahmin işlemi için CNN, LSTM ve GRU gibi derin öğrenme modellerinin kullanımı ile birlikte bu modellerin birleştirilmesi üzerine yeni bir hibrit model üzerine odaklanılmıştır. Her bir algoritma, geçmiş gümüş fiyat verileri üzerinde eğitilmiş ve bu verileri kullanarak fiyat tahminlerindeki performansları karşılaştırılmıştır. Bu yaklaşım, her bir modelin güçlü yönlerini birleştirerek daha kapsamlı ve hassas tahminler elde etmeyi hedefler. Ayrıca, finansal tahminlerde sıklıkla göz ardı edilen gümüş piyasası gibi özel bir alanı ele alarak, bu alandaki literatüre özgün bir katkı sağlamaktadır. Çalışma, zaman serisi verilerinin işlenmesi konusunda derin öğrenme modellerinin avantajlarını ve potansiyelini vurgulayarak, finansal tahmin ve analiz metodolojilerinde yenilikçi bir yaklaşım sunmaktadır. Sonuçlar, bu algoritmaların sadece geçmiş verilere dayalı olarak gümüş fiyatlarını analiz etme ve geçmiş trendleri değerlendirme yeteneklerini karşılaştırmıştır. Çalışma, bu algoritmaların geçmiş verilere dayalı analizlerde farklı performanslar sergilediğini göstermiştir. Sonuç olarak, bu çalışma, gümüş fiyatlarının geçmiş verileri üzerinden tahmin edilmesi için farklı derin öğrenme algoritmalarının performansını karşılaştırmış ve CNN-LSTM-GRU hibrit modelinin daha iyi tahminler yapma potansiyeli taşıdığını ortaya koymuştur. Bu sonuçlar, finansal analiz ve tahmin konularında çalışan araştırmacılara yol gösterici olabilir.

Keywords

Gümüş Fiyat Tahmini, Derin Öğrenme, LSTM, CNN-LSTM-GRU, Finansal Analiz

Ethical Statement

Bu çalışma için etik kurul iznine ihtiyaç duyulmamıştır.

Comparative Analysis of Deep Learning Models for Silver Price Prediction: CNN, LSTM, GRU and Hybrid Approach

Abstract

In this study, the performance of different deep learning algorithms to predict silver prices was evaluated. It was focused on the use of deep learning models such as CNN, LSTM, and GRU for the prediction process, as well as a new hybrid model based on combining these models. Each algorithm was trained on historical silver price data and compared its performance in price prediction using this data. This approach aims to achieve more comprehensive and accurate forecasts by combining the strengths of each model. It also makes a unique contribution to the literature in this area by addressing a specialized area such as the silver market, which is often neglected in financial forecasting. The study presents an innovative approach to financial forecasting and analysis methodologies, highlighting the advantages and potential of deep learning models for time-series data processing. The results compare the ability of these algorithms to analyze silver prices based on historical data only and to assess past trends. The study showed that these algorithms exhibit different performances in analyzing historical data. In conclusion, this study compared the performance of different deep learning algorithms for predicting silver prices based on historical data and found that the CNN-LSTM-GRU hybrid model has the potential to make better predictions. These results can provide guidance to researchers working on financial analysis and forecasting.

Keywords

Silver Price Prediction, Deep Learning, LSTM, CNN-LSTM-GRU, Financial Analysis

References

• Alshaikhdeeb, A. J. & Cheah, Y. N. (2023). Utilizing Word Index Approach with LSTM Architecture for Extracting Adverse Drug Reaction from Medical Reviews. Journal of Advances in Information Technology, 14(3).
• Ayzel, G., & Heistermann, M. (2021). The effect of calibration data length on the performance of a conceptual hydrological model versus LSTM and GRU: A case study for six basins from the CAMELS dataset. Computers & Geosciences, 149, 104708.
• Brownlee, J. (2020), How to Grid Search Deep Learning Models for Time Series Forecasting, https://machinelearningmastery.com/how-to-grid-search-deep-learning-models-for-time-series-forecasting/ Access Date: 18.12.2023
• Buslim, N., Rahmatullah, I. L., Setyawan, B. A., & Alamsyah, A. (2021, September). Comparing Bitcoin's Prediction Model Using GRU, RNN, and LSTM by Hyperparameter Optimization Grid Search and Random Search. In 2021 9th International Conference on Cyber and IT Service Management (CITSM) (pp. 1-6). IEEE.
• Cansu, T., Kolemen, E., Karahasan, Ö., Bas, E., & Egrioglu, E. (2023). A new training algorithm for long short-term memory artificial neural network based on particle swarm optimization. Granular Computing, 1-14.
• Chen, J. (2023). Analysis of bitcoin price prediction using machine learning. Journal of Risk and Financial Management, 16(1), 51.
• Cho K., Van Merrienboer B., Gulcehre C.et al., (2014), Learning phrase representations using RNN encoder-decoder for statistical machine translation, arXiv preprint arXiv:1406.1078, 2014.
• Doke, P., Shrivastava, D., Pan, C., Zhou, Q., & Zhang, Y. D. (2020). Using CNN with Bayesian optimization to identify cerebral micro-bleeds. Machine Vision and Applications, 31, 1-14.
• Gao, Y., Wang, R. & Zhou, E. (2021). Stock Prediction Based on Optimized LSTM and GRU Models. Scientific Programming, 2021. https://doi.org/10.1155/2021/4055281
• Goel, S., Saxena, M., Sarangi, P. K. & Rani, L. (2022). Gold and Silver Price Prediction using Hybrid Machine Learning Models. In 2022 Seventh International Conference on Parallel, Distributed and Grid Computing (PDGC), (pp. 390-395). IEEE.
• Hamayel, M. J. & Owda, A. Y. (2021). A novel cryptocurrency price prediction model using GRU, LSTM and bi-LSTM machine learning algorithms. AI, 2(4), 477-496.
• Hsieh, C. H., Li, Y. S., Hwang, B. J. & Hsiao, C. H. (2020). Detection of atrial fibrillation using 1D convolutional neural network. Sensors, 20(7), 2136.
• Investing.com Data: https://tr.investing.com/currencies/xagg-try-historical-data Access Date: 30.09.2023.
• Kong, D., Liu, S. & Pan, L. (2021). Amazon spot instance price prediction with GRU network. In 2021 IEEE 24th international conference on computer supported cooperative work in design (CSCWD), (pp. 31-36). IEEE.
• Kumar, G., Singh, U. P., & Jain, S. (2022). An adaptive particle swarm optimization-based hybrid long short-term memory model for stock price time series forecasting. Soft Computing, 26(22), 12115-12135.
• Lara-Benítez, P., Carranza-García, M., & Riquelme, J. C. (2021). An experimental review on deep learning architectures for time series forecasting. International journal of neural systems, 31(03), 2130001.
• Li, F., Zhou, H., Liu, M. & Ding, L. (2023). A Medium to Long-term Multi-influencing Factor Copper Price Prediction Method Based on CNN-LSTM. IEEE Access, (99), 1-1.
• Lin, Y., Liao, Q., Lin, Z., Tan, B. & Yu, Y. (2022). A novel hybrid model integrating modified ensemble empirical mode decomposition and LSTM neural network for multi-step precious metal prices prediction. Resources Policy, 78, 102884.
• Malik, A., Gupta, P. & Vijh, S. (2022). Towards a Stock Price Prediction on Time Series Data using Long-Short Term Memory Method. In 2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence), (pp. 525-529). IEEE.
• Patel, N. P., Parekh, R., Thakkar, N., Gupta, R., Tanwar, S., Sharma, G., ... & Sharma, R. (2022). Fusion in cryptocurrency price prediction: A decade survey on recent advancements, architecture, and potential future directions. IEEE Access, 10, 34511-34538.
• Pranolo, A., Mao, Y., Wibawa, A. P., Utama, A. B. P., & Dwiyanto, F. A. (2022). Robust LSTM With tuned-PSO and bifold-attention mechanism for analyzing multivariate time-series. IEEE Access, 10, 78423-78434.
• Rao, B. S., Bhattacharya, R., Tiwari, M. K., Kumari, K. A., Devmane, M. A. & Singh, K. (2023). Innovative Deep Learning Model-based Stock Price Prediction using a Hybrid Approach of CNN and Gradient Recurrent Unit. In 2023 8th International Conference on Communication and Electronics Systems (ICCES), (pp. 1304-1309). IEEE.
• Sbrana, G., & Silvestrini, A. (2022). Random coefficient state-space model: Estimation and performance in M3–M4 competitions. International Journal of Forecasting, 38(1), 352-366.
• Sulistio, B., Warnars, H. L. H. S., Gaol, F. L. & Soewito, B. (2023). Energy Sector Stock Price Prediction Using The CNN, GRU & LSTM Hybrid Algorithm. In 2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE), (pp. 178-182). IEEE.
• Vidya, G. S. & Hari, V. S. (2020). Gold price prediction and modelling using deep learning techniques. In 2020 IEEE Recent Advances in Intelligent Computational Systems (RAICS), (pp. 28-31). IEEE.
• Vrigazova, B. (2021), “The Proportion for Splitting Data into Training and Test Set for the Bootstrap in Classification Problems”, Business Systems Research, 12(1):228-242. DOI: https://doi.org/10.2478/bsrj-2021-0015
• Wang, H., Dai, B., Li, X., Yu, N. & Wang, J. (2023). A Novel Hybrid Model of CNN-SA-NGU for Silver Closing Price Prediction. Processes, 11(3), 862.
• Wibawa, A. P., Utama, A. B. P., Elmunsyah, H., Pujianto, U., Dwiyanto, F. A., & Hernandez, L. (2022). Time-series analysis with smoothed Convolutional Neural Network. Journal of big Data, 9(1), 44.
• Xu, Y., Hu, C., Wu, Q., Jian, S., Li, Z., Chen, Y., ... & Wang, S. (2022). Research on particle swarm optimization in LSTM neural networks for rainfall-runoff simulation. Journal of hydrology, 608, 127553.
• Yang, J., De Montigny, D. & Treleaven, P. (2022, May). ANN, LSTM, and SVR for gold price forecasting. In 2022 IEEE Symposium on Computational Intelligence for Financial Engineering and Economics (CIFEr), (pp. 1-7). IEEE.
• Zulfiqar, M., Gamage, K. A., Kamran, M., & Rasheed, M. B. (2022). Hyperparameter optimization of bayesian neural network using bayesian optimization and intelligent feature engineering for load forecasting. Sensors, 22(12), 4446.