Predicting the stock market using machine learning algorithms: evidence from India and Brazil

Artificial Intelligence and Machine Learning

This study investigates the use of machine learning algorithms to predict stock market movements, focusing on the National Stock Exchange of India and Bolsa de Valores de So Paulo of Brazil. The research contrasts six algorithms—Random Forest, Support Vector Machine, Decision Tree, Multilayer Perceptron, K-Nearest Neighbours, and Clustering— against individual and combined market data. Random Forest outperforms other methods across datasets in terms of accuracy and error rates. A comparative examination reveals that algorithmic performance and market dynamics are expanding. The research also considers the implications for regulatory agencies, clients, academics, and industry counterparts, providing a multidimensional perspective of machine learning's potential and concerns in stock market forecasting. Despite positive results, the study cautions against overfitting and market and algorithm specialisation. According to the dissertation, future research should diversify markets and algorithms, integrate real-time data, develop hybrid models, and address ethical concerns. These findings advance financial analytics and lay the foundation for machine learning in financial markets.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.