Supervised decision making in forex investment using ML and DL classification methods
Is this your thesis?
This record came from a bulk archive import. If it’s yours, link it to your profile.
Abstract (EN)
The foreign exchange (FX) market, characterized by its complexity and high trad- ing volumes, presents a challenging environment for traders, who must navigate vast amounts of data and make rapid decisions under high-pressure conditions. This study proposes a novel, comprehensive, and realistic trading system for the GBP/JPY currency pair, integrating machine learning and deep learning tech- niques to address the challenges faced by traders in the financial markets. The proposed algorithm combines information from different timeframes, predicts in- dicator values using the Long Short-Term Memory (LSTM) model, and employs the XGBoost classifier for the final buy or sell decision. The research employs a sliding window training approach to adapt to the rapidly changing patterns in the forex market, enabling the detection of subtle variations in price dynamics. The experiment, conducted over three consecutive months, demonstrates promising results in terms of capital growth and prediction ac- curacy. However, the method is considered high-risk from a risk management perspective, necessitating further refinement through the incorporation of risk management techniques and algorithm optimization. This study contributes to the advancement of trading strategies by bridging the gap between researchers and traders, showcasing the potential of machine learn- ing and deep learning techniques to enhance decision-making processes in the financial markets. The proposed trading system paves the way for the develop- ment of more robust, automated trading systems, ultimately empowering traders to navigate the complexities of the ever-evolving foreign exchange market with greater confidence and precision.
Author
Abdullah Jıroudı
Institution
How to Cite
Abdullah Jıroudı (Master Thesis). Supervised decision making in forex investment using ML and DL classification methods, 2023, Feyziye Schools Foundation Işık University.
Keywords
License
Tüm Hakları Saklıdır
This work is shared under the specified license terms.
More theses from Feyziye Schools Foundation Işık University
- Examining the role of psychological well-being and sexual satisfaction in predicting marital adjustment(2021)
- Relations of 1994 and 2000 crises and their effects in Turkish capital markets(2002)
- Design of reed-solomon and convolutional coded OFDM system and performance analysis(2003)
- Evaluation of innovativeness of Turkey with respect to european union integration(2003)
- Project workspace: A computer supported collorative work tool for software project development(2005)
- Performance analysis of static agent architecture with centralized matchmaking(2005)
