Stock Price Movement Prediction using Machine Learning Algorithms and Time Series Models: A Review

Irfan Ramzan  Parray

Abstract

The stock market is a very important place for investment in a country. It also serves as the index of the growth for an economy. To determine the behavior of the stock market trends has been the focal point of researchers for very long. The nonlinear structure of the stock market makes it challenging to predict how the stock price will change. But it has been demonstrated that stock market behavior can be predicted using various machine learning algorithms. These techniques use historical data of the stocks to train the machine learning algorithm and predict their future behavior. Also there exist some time series models that can help in forecasting the stock trend since the historical data of the stock market is time series data. This study reviews the various machine learning algorithms and time series models that are used for stock market predictions. It also implements Long Short-Term Memory (LSTM) and Support Vector Regression (SVR) algorithms to predict the stock market behavior.

Abstract

Keywords