Comparative Analysis between Supervised Machine Learning and Time Series Models for stock price prediction

Share markets are considered as an important part of a country’s economy and efficiently functioning stock markets give companies the ability to quickly access capital from the public. This paper aims at identifying the best potential method to predict the adjusted close prices of stocks, through a comparative analysis among Supervised Learning and Time Series models. Techniques like Sliding Window, Forward Fill and Linear Interpolation have been used to preprocess the data. Predictions using Linear Regression, Random Forests, Support Vector Machines, Auto Regressive Integrated Moving Average and Auto Regressive Integrated Moving Average with Exogenous variables, were performed and evaluated based on their prediction accuracies. After the individual evaluation of all the methods, Support Vector Machines are found to give better results for stock price prediction.