Prediction of stock returns in Istanbul stock exchange using machine learning methods

The stock market parameter forecasting is an important research subject both for financial professionals and the machine learning experts due to the challenges and opportunities it possess. Despite the difficulties in financial data, interest in this research area is growing rapidly. In this study we examined the data of the 25 leading companies of Borsa Istanbul and applied several estimation algorithms. As a result, we observed that the Random Forest algorithm gave the best result with an accuracy of 57.37%.