Analyzing fluctuations of stock prices of the companies listed in Tehran Stock Exchange Using the machine learning method

The changes in stock prices result in volatility in returns of companies. Hence, these changes are stimulating issues for investors. The purpose of this research is to predict the fluctuation of stock prices in the Iranian stock market using the machine learning algorithm. In this two-step research, firstly, 31 variables influencing stock price volatility were extracted from the financial information of 1750 years of the company. Then,  stock price volatility was predicted in terms of selected variables using MATLAB software. The findings show that nonlinear communication algorithm using accounting variables has higher power in predicting stock price volatility than linear communication algorithm. Therefore, it is recommended that shareholders and corporates’ decision makers rely on the predictive power of artificial intelligence algorithms, especially the nonlinear mode of relevance vector machine (RVM), in decision making on investment in the stock market