Title: Analysis of machine learning's performance in stock market prediction, compared to traditional technical analysis indicators

Authors: Mohammed Bouasabah

Addresses: Department of Business and Management, National School of Business and Management, Ibn Tofail University, Kenitra, Morocco

Abstract: This study compares the performance of machine learning (ML) algorithms with traditional technical indicators in real estate, technology, and healthcare sectors. Unveiling the limitations of classical indicators, particularly their struggle to surpass the 50% threshold, the research explores the predictive capabilities of ML algorithms, focusing on AdaBoost and support vector machine (SVM). The relative strength index (RSI) emerges as a reliable performer for buy decisions but with potential oversight. Results affirm the superiority of ML algorithms in precision, recall, and F1 score, transcending traditional indicators. Sector-specific variations showcase exceptional ML efficacy, particularly in healthcare. Algorithmic evaluation spotlights AdaBoost and SVM, underscoring the importance of strategic selection. The study advocates for a nuanced approach, blending RSI with ML for refined strategies. In conclusion, this research contributes significantly to financial decision-making, exposing limitations and positioning ML algorithms as powerful tools for improved investment strategies.

Keywords: machine learning; ML; technical analysis indicators; prediction; financial market; data analysis; SMA; MACD; RSI; trading; support vector machine; SVM.

DOI: 10.1504/IJDATS.2024.137465

International Journal of Data Analysis Techniques and Strategies, 2024 Vol.16 No.1, pp.32 - 46

Received: 08 Jul 2023
Accepted: 15 Dec 2023
Published online: 19 Mar 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article