Using Brownian motion in stock prices prediction in comparison with ARIMA

The main reason that people invest in the stock market is to earn profits that require having accurate market information and stock changes and predicting its future trend. Therefore, the investor needs the powerful and reliable tools needed to predict stock prices. In this regard, the present study investigates stock price forecasts based on MSE mean square error, mean absolute deviation MAE and root mean square error RMSE. Finally, the methods investigated in this study are compared and identify the top method to predict stock prices. For this purpose, the data of the top 50 stock exchange companies, which are quarterly presented by the stock exchange organization, were used during the period 2012-2018. In order to test the research hypotheses, linear regression method, Brownian method and ARIMA method were used. The research findings show that the Brownian model predicts stock prices more accurately than the ARIMA method. It was also observed that linear statistical ARIMA models are less efficient in the financial markets than the brownian methods.