A machine learning approach to predicting stock returns

Abstract

Machine learning approaches to stock market forecasting have become increasingly popular throughout the years due to their predictive power and ability to identify hidden patterns in the data. However, considering the inherent volatility and complexity of stock markets, this is a challenging problem to model. This paper presents a comparative analysis of the performance of various machine learning regression algorithms in predicting stock returns. Several leading and technical indicators are considered as features to predict the monthly return of the S&P 500 Index, a market-capitalization-weighted index of the 500 largest publicly traded companies in the United States.