Perpustakaan Digital ITB

Technical analysis is one way to predict future stock prices by observing the trends of data time series in graphical form. Besides that, investors need a long time periode to observe and to analyze the data time series. This study developed a system to answer this problem by using ANFIS and JADE (Java Agent Development Framework)-based agent applications. The ANFIS is used to predict future stock prices and the JADE-based agent application is used to retrieve data from a website and it runs the system automatically to provide the results of the stock price prediction. The accuracy of this system that uses ANFIS in this study is 88.25% for real data stock and 85.17% for simulation data. In addition, JADE-based agent has been run well for retrieving data from the website and it has run the system automatically to display the results of prediction for the future stock price. Therefore, this system can minimalsize the time for investors to predict stock price in the future.