The Integration of machine based learning into technical analysis of financial markets

Ing. Yosra Touirtou

Diplomová práce

Abstract:

This thesis explores the integration of machine learning techniques with technical analysis to enhance the forecasting accuracy of the EUR/USD exchange rate. The study provides insights into whether combining machine learning algorithms with technical analysis indicators such as trend following, momentum and volatility indicators can improve the prediction accuracy of the foreign exchange financial market. The research investigates the predictive performance of linear regression, random forest and XG boost while using a data set of hourly observations of EUR/USD prices from 2010 to the beginning of 2025. The results reveal that nonlinear models (Xgboost, Random Forest) significantly outperform the predictive performance of linear regression. The nonlinear models show better adaptability to the complexity of the time series data. Among these, the Random Forest model demonstrates the strongest and most consistent performance, while Xgboost achieves comparably strong results. The ensemble models combining the predictions from Random Forest and XGboost shows broader market dynamics yet lower forecasting accuracy. The models perform reliably during stable market conditions. However, their accuracy decreases in the case of volatility peaks and unforeseen macroeconomic events. Overall, the results entitle the effectiveness in integrating machine learning tools with technical analysis indicators to improve the predictive performance of the foreign exchange market, thereby supporting the decision-making processes for traders and investors.

Keywords

Machine learning, Technical Analysis, Forex market, Fianacial time series forecasting, Non linear Models

Jazyk práce: angličtina

Datum vytvoření / odevzdání či podání práce: 9. 5. 2025

Obhajoba závěrečné práce

• Obhajoba proběhla 17. 6. 2025
• Vedoucí: Ing. Michala Moravcová, Ph.D.
• Oponent: prof. Ing. Štefan Lyócsa, PhD.

Masarykova univerzita

Ekonomicko-správní fakulta

Magisterský studijní program / obor:
Finance / Finance