Predicting the Direction of Movement for Stock Price Index Using Machine Learning Methods

Abstract

Stock price prediction with high accuracy may offer significant opportunities for the investors who make decisions on making profit or having high gains over the stocks in stock markets. The aim of this study is to predict the movement directions (UP/DOWN) of the Istanbul Stock Exchange National 100 (ISE National 100) Index accurately for short-term futures by using three machine learning methods, which are Logistic Regression (LR), Support Vector Machines (SVMs), and Multilayer Perceptron (MLP). Two datasets used in this study are composed of sessional and daily points of data over a 5-year period from November 2007 to November 2012. During the prediction of the movement directions, the following factors were taken into account; data of macroeconomic indicators, gold prices, oil prices, foreign exchange prices, stock price indexes in various countries, and the data of ISE National 100 index for past sessions and prior days, which are used as input variables in the datasets. In connection with that, the most effective features of these input variables were determined by using some feature selection methods. As a result, the movement directions of ISE National 100 were predicted with higher accuracies by using reduced datasets than original datasets and the best performances were found by LR classifier.