A Study on Machine Learning Algorithms for Indian Stock Market Predictions

Abstract

Because of its inherent complexity and variety, the stock market has always piqued the interest of those seeking methods to foretell its future movement. But this is made more challenging by the fact that financial markets throughout the world are quite volatile. While statistical approaches and modeling do a good job, they aren't foolproof when it comes to predicting the stock market's behavior. Addressing and solving the complex challenges in the stock market was beyond the capabilities of the old approaches utilized. Machine learning and artificial intelligence solutions simplify the handling of Big Data's complicated scenarios and difficulties. The goal of this study is to find a model that can predict outcomes that are close to the real ones by combining six distinct algorithms: Support Vector Machine, Generalized Linear Model, Deep Learning, Decision Tree, Random Forest, and Gradient Boosted Trees. The BSE index data from April 2015 to March 31, 2020 was processed using these techniques, and the model with the least relative error was determined. Gradient Boosted Trees is selected as the most efficient model out of all the ones used since it has the lowest relative error and standard deviation. The outcomes are predicted using Additional Gradient Boosted Trees.