Stock index reflects the price movement a group of stock. There are many stock indices in the world. JKSE (Jakarta Composite Index) is one of the stock indices in IDX (Indonesia Stock Exchange). There are many benefits in knowing the movement of JKSE value, one of them is to reduce the risk in investing in the stock market. Therefore it is a need to predict JKSE value. The method that is used in this paper is ANFIS (Adaptive Neuro Fuzzy Inference System). ANFIS is a hybrid model which can give the better accuracy than other isolated technique of AI (Artificial intelligence) such as ANN, fuzzy logic, and GA. This paper gives two outputs, they are the prediction result and classification result based on some relative error values. Classification method that is used in this paper is Fuzzy Kernel C-Means. This two outputs will help the investors in decision making. The experimental results give an average of prediction accuracy 91 % and classification accuracy 80.2 %.