Predict the trend of stock prices using machine learning techniques
Keywords:
technical indicators, support vector machine, random forest and k-nearest neighbourAbstract
According to growing importance of the stock market in economic conditions of each country and since, stock prices are the most important factors influencing investment decisions for selecting stock, predicting the stock price movement is an integral part of the investment. This paper presented to forecast the movement of stock prices Tejarat bank of Iran with considerable precision. Accordingly, to predict the trend of stock prices using machine learning techniques and economic indicators have been considered. About 18,000 different indicators are presented, both simple moving average (SMA), weighted moving average (WMA), relative strength indicator (RSI) and moving average convergence divergence (MACD) indicator those are widely used in the stock market of Iran, have been chosen. The output of those is input three classifier, support vector machines, random forests and k-nearest neighbour. The outputs of the three classifier will be compared with each other. The results in this paper show that, respectively, random forest classifier, support vector machine and the k- nearest neighbours have the best accuracy in categories.
