An Analytical Study on Model of Stock Index prediction Based on Data Mining

Numerous individuals consider stock market prediction as gambling. Anyway it is conceivable to create helpful patterns by the investigation of stock prices. Data mining systems can be connected on over a wide span of time budgetary data to create examples and decision making algorithms. Prediction of stock market list is testing errand investors to contribute the money. The primary targets of investors are contributing the money and make profit. The point of this paper is to build up the robust predictive model which predicts the right foresting with minimum error. The instruments utilized in this investigation are machine learning techniques like Classification and Regression Technique (CART), CHAID, Artificial Neural Network (ANN) and Support Vector Machine (SVM) for examination and prediction of BSE SENSEX data.. Anticipating stock return is a vital financial subject that has pulled in analysts' consideration for a long time The ANN gives the better prediction with extremely less Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). We have additionally expand the experimental work and broke down the ANN predictive model with various learning rate and accomplished less error estimates like MAE =0.0044 and MAPE= 0.676 with 0.9 learning rate and 1 hidden layer This think about attempts to help the speculators in the stock market to choose the better planning for purchasing or selling stocks dependent on the information separated from the prices of such stocks.