Predicting the direction of Indonesian stock price movement using support vector machines and fuzzy Kernel C-Means

摘要: The nature of stock price fluctuations becomes a challenge for the investors to gain return in investing stocks. To overcome this problem, need some accurate predictions order anticipate future movement. However, predicting direction movement is complex task due many uncertain factors affecting itself. Therefore, paper studied application Support Vector Machines and Fuzzy Kernel C-Means Indonesian market, particularly on banking subsector. Using historical data, eight technical indicators have been computed obtain two different approaches input model. One them use while other process into trends. results suggest that, general view, with represented as trend being model outperforms prediction models. particular condition, best entire observation 92 % accuracy given by FKCM using σ = 100 90 training data.