|
Design of a Hybrid Algorithmic Trading System Optimized by Genetic Algorithm
|
Shiva Ghasempoura1   , Amir Reza Keyghobadib *2 , Shadi Shahverdiani3 , Mahdi Madanchi Zajd4 |
1- a Department of Financial Management, SR.C., Islamic Azad University, Tehran, Iran
2- Department of Industrial Management, CT.C., Islamic Azad University, Tehran, Iran & a Department of Financial Management, SR.C., Islamic Azad University, Tehran, Iran , ar.keyghobadi@iau.ac.ir
3- Department of Business Administration, ShQ.C., Islamic Azad University, Shahr-e Qods, Iran & a Department of Financial Management, SR.C., Islamic Azad University, Tehran, Iran
4- d Department of Financial Management, CT.C., Islamic Azad University, Tehran, Iran. & a Department of Financial Management, SR.C., Islamic Azad University, Tehran, Iran |
|
|
Abstract: (6 Views) |
In recent years, the use of algorithmic trading systems has significantly increased in global financial markets and, more recently, in Iran’s capital market. The main advantage of such systems lies in enhancing the speed and accuracy of trading decisions while eliminating emotional interference. This study aims to design an optimized hybrid algorithmic trading system based on a genetic algorithm. A collective learning approach was employed by integrating four classification algorithms—Decision Tree (DTree), K-Nearest Neighbor (KNN), RUSBoost, and Radial Basis Neural Network (RBN)—to simultaneously achieve optimal feature selection, parameter tuning, and weight determination for each algorithm. The research data include price information and technical indicators of selected stocks listed on the Tehran Stock Exchange during 2012–2021. The genetic algorithm was used to optimize 51 decision variables, including algorithm parameters, selected features, and model weights. The results demonstrate that the proposed hybrid model achieves high classification accuracy in identifying buy, sell, and hold classes, outperforming single models. Therefore, the developed system can serve as an effective and optimized automated trading framework for Iran’s capital market. |
Article number: 21 |
| Keywords: Algorithmic trading, Genetic algorithm, Ensemble learning, Optimization, Neural networks |
|
|
Full-Text [PDF 505 kb]
(2 Downloads)
|
Article type: Research |
Subject:
Special
Received: 2025/10/26 | Accepted: 2025/12/16 | Published: 2026/08/1
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
