Ανάπτυξη στρατηγικής αλγοριθμικής διαπραγμάτευσης με την χρήση μηχανικής μάθησης

Νικητάκης, Εμμανουήλ

Development of an algorithmic trading strategy using machine learning

Bachelor Dissertation

Author

Νικητάκης, Εμμανουήλ

Date

2025-09

Abstract

This thesis examines the role of machine learning in the development of algorithmic trading techniques and analyzes how this combination decisively influences the overall behavior of financial markets. It begins with an extensive literature review, presenting the fundamental concepts of algorithmic trading and intelligent techniques, the different types of markets, as well as the various financial terms related to trading activities. At the same time, the main challenges arising during the implementation of automated strategies are discussed, along with the existing operational frameworks and control mechanisms developed by regulatory authorities worldwide to ensure transparency and stability. The research contribution of this thesis focuses on technical analysis and, more specifically, on the design and implementation of an automated signal-based strategy based on the logic of trend following and mean reversion, applied to four different financial assets. Subsequently, breakout signals combined with a technical indicator are used as inputs for training an XGBoost classifier model, aiming to predict the future market direction. The results obtained from these four financial assets, both from the automated algorithmic strategy and the machine learning model, were evaluated through charts, comparative tables, and various performance metrics.

Department

Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Πληροφορικής

Number of pages

Language

Greek

URI

https://dione.lib.unipi.gr/xmlui/handle/unipi/18146

Collections

Τμήμα Πληροφορικής

Show full item record

Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα