Forecasting stock market trend: a comparison of machine learning algorithms

Abstract

Forecasting the direction of stocks markets has become a popular research topic in recent years. Differentapproaches have been applied by researchers to address the prediction of market trends by consideringtechnical indicators and chart patterns from technical analysis. This paper compares the performanceof four machine learning algorithms to validate the forecasting ability of popular technical indicators inthe technological NASDAQ index. Since the mathematical formulas used in the calculation of technicalindicators comprise historical prices they will be related to the past trend of the market. We assume thatforecasting performance increases when the trend is computed on a longer time horizon. Our resultssuggest that the random forest outperforms the other machine learning algorithms considered in ourresearch, being able to forecast the 10-days ahead market trend, with an average accuracy of 80%.