Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68622
Volume:
10 |
10 |
Oct 2026
Pages:
1-5
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
AIML
Zeal Polytechnic Pune
AIML
Zeal Polytechnic Pune
AIML
Zeal Polytechnic Pune
AIML
Zeal Polytechnic Pune
AIML
Zeal Polytechnic Pune
AIML
Zeal Polytechnic Pune
The Foreign Exchange (Forex) market is a global, decentralized financial market where currencies are traded in pairs. Predicting currency price movements is exceptionally challenging because exchange rates are influenced by dynamic, complex factors including macroeconomic indicators, central bank policies, market volatility, geopolitical events, and investor sentiment. Traditional technical analysis relies on indicators such as Moving Average (MA), Relative Strength Index (RSI), and Moving Average Convergence Divergence (MACD) to evaluate historical price trends and momentum. However, these individual indicators often produce conflicting or lagging signals under highly turbulent, non-stationary market conditions. Machine Learning (ML) and Deep Learning (DL) models offer robust data-driven alternatives by autonomously learning intricate linear and non-linear patterns from time-series historical data. This paper presents a comprehensive, structured survey of state-of-the-art machine learning and deep learning techniques applied to Forex market forecasting and algorithmic trading strategy development. It systematically reviews supervised learning paradigms including Support Vector Machines (SVM), Decision Trees, Random Forest (RF), and Artificial Neural Networks (ANN), alongside modern recurrent architectures designed specifically for temporal sequences, such as Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and hybrid GRU-LSTM frameworks. Furthermore, this study examines feature engineering protocols, technical indicator integration, model evaluation metrics (MAE, MSE, RMSE, Directional Accuracy, Sharpe Ratio), and time-series cross-validation schemes (such as walk-forward validation) necessary to prevent severe data leakage. Finally, key operational challenges—including non-stationarity, extreme market volatility, noise, overfitting, order execution latency, and realistic transaction costs (spreads, slippage, and overnight fees)—are analyzed, highlighting promising prospective research directions such as multi-task technical indicator prediction and explainable financial AI.
KEYWORDS
This paper presented a detailed, rigorous survey on machine learning and deep learning techniques applied to Forex trading prediction and analysis. It reviewed technical indicators (MA, RSI, MACD, Bollinger Bands), supervised learning algorithms (SVM, Random Forest, ANN), and advanced temporal architectures (LSTM, GRU, hybrid GRU-LSTM networks). The survey highlighted the critical necessity of proper chronological walk-forward validation to eliminate data leakage and emphasized the imperative of incorporating realistic transaction costs into quantitative backtests. By identifying key operational limitations and proposing novel multi-task technical indicator prediction paradigms, this study provides a solid foundation and clear roadmap for researchers and quantitative practitioners developing robust, automated FX trading intelligence systems.
[1] M. Ayitey Junior, P. Appiahene, O. Appiah, and C. N. Bombie, "Forex market forecasting using machine learning: Systematic literature review and meta-analysis," Journal of Big Data, vol. 10, art. 9, 2023. doi: 10.1186/s40537-022-00676-2. [2] A. Santuci, E. Sbruzzi, L. Araújo-Filho, and M. Leles, "Evaluation of FOREX trading strategies based on Random Forest and Support Vector Machines," IEEE Latin America Transactions, vol. 20, no. 9, pp. 2146–2152, 2022. doi: 10.1109/TLA.2022.9878170. [3] P. Escudero, W. Alcocer, and J. Paredes, "Recurrent neural networks and ARIMA models for Euro/Dollar exchange rate forecasting," Applied Sciences, vol. 11, no. 12, art. 5658, 2021. doi: 10.3390/app11125658. [4] M. S. Islam and E. Hossain, "Foreign exchange currency rate prediction using a GRU-LSTM hybrid network," Soft Computing Letters, vol. 3, art. 100009, 2021. doi: 10.1016/j.socl.2020.100009. [5] Y. Fischer and C. Krauss, "Deep learning with long short-term memory networks for financial market predictions," European Journal of Operational Research, vol. 270, no. 2, pp. 654–669, 2018. [6] J. B. Heaton, N. G. Polson, and J. H. Witte, "Deep learning in finance," Applied Stochastic Models in Business and Industry, vol. 33, no. 1, pp. 3–12, 2017. [7] A. A. Ariyo, A. O. Adewumi, and C. K. Ayo, "Stock price prediction using the ARIMA model," in Proc. UKSim-AMSS 16th Int. Conf. Comput. Model. Simul., 2014, pp. 106–112. [8] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. [9] K. Cho, B. van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, "Learning phrase representations using RNN encoder-decoder for statistical machine translation," in Proc. EMNLP, 2014, pp. 1724–1734.
Publication Date
10/31/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
4
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...