مجلة التنمية الاقتصادية
Volume 11, Numéro 1, Pages 355-368
2026-06-30
Machine Learning Applications In Predicting Stock Returns: A Scientific And Literary Approach To Modern Methods
Authors : Bentireche Atallah .
Abstract
Summary: The global financial landscape is undergoing a structural shift driven by digital transformation and the integration of high-velocity big data. Traditional econometric models are increasingly supplemented or replaced by machine learning (ML) architectures that can process non-linear relationships. Despite the proliferation of AI, economic actors suffer from "bounded rationality," where cognitive and informational constraints hinder optimal decision-making. Furthermore, market volatility and the "black box" nature of advanced algorithms create risks for institutional and retail investors. This study aims to analyze the efficacy of modern ML methods—such as Random Forest, XGBoost, and Deep Neural Networks—in predicting stock returns and identifying market anomalies while evaluating their impact on financial market efficiency. Drawing on a multi-dimensional bibliometric analysis and comparative empirical studies across diverse markets (US, China, South Africa, and Algeria), the research evaluates performance through metrics like Accuracy, F1 Score, and the Matthews Correlation Coefficient (MCC). Findings indicate that ensemble methods like Random Forest and gradient-boosted models consistently outperform linear regressions by 5–15% in predictive accuracy. Additionally, deep learning models achieve up to 93.8% accuracy in stock classification by incorporating financial ratios from 10-K reports. This research bridges the gap between behavioral economics (bounded rationality) and computational finance, proposing a hybrid framework for "dynamic efficiency" in modern markets
Keywords
Machine Learning ; Stock Returns ; Big Data Analytics ; Algorithmic Trading ; Financial Transparency ; Bounded Rationality
Les articles similaires
جغدير الطاهر
.
صاطوري الجودي
.
ص 985-1004.
Gharbi Hamza
.
Bedrouni Aissa
.
pages 19-34.
Bouriche Nouria
.
Benbouziane Mohaned
.
pages 462-473.
Bagheffar Abdelkader
.
Saous Cheikh
.
pages 166-183.
Elgabali Dr Walid
.
pages 205-225.
