会计与营销

Predicting Trend of Stock Prices by Developing Data Mining Techniques with the Aim of Gaining Profit

Abstract

Seyed Saman Emami

In financial markets, for the purpose of forecasting the future trend of stock prices, technical analysis tools are used, with the tools of the analysis including harmonic patterns, time tools and technical indicators. Due to the fact that technical indicators play a very important role in predicting the future trend, they also confirmed the results of other technical models and instruments, so these indicators have a great importance in technical analysis. Traders of financial markets are faced with a wide range of these technical indicators, so that the accuracy of each of them is not specified, as well as many traders which used them tactfully and effortlessly, so, in this study, using individual machine learning algorithms, we determined the significance of each of these indicators. Also, the ensemble learning algorithm has been developed with the goal of increasing predictive efficiency. In the forthcoming study, before entering into the machine learning discussion, 12 popular indicators are ranked by market participants by the VIKOR method. Also, with using Decision Tree, Artificial Neural Network, Support Vector Machine and Logistic Regression, the importance of each indicator based on precision Specified nose and output values   have been calculated. Further, the ensemble learning model has been developed with the aim of increasing the efficiency and accuracy by the optimization technique and the results of this model have been compared with the Weighted Majority Voting method. The results showed that OBV, CCI and EMA indicators are very important. Also, the results of machine learning indicated the superiority of the results of the SVM method in prediction accuracy.