Indian Stock Market Prediction Based on Rough Set and Support Vector Machine Approach

Abstract

In recent trends, the stock market prediction has been treated as a challenging task for every people who have been associated with the financial market. Forecasting is the activity of historic data to set up the direction of future trends. It is a very difficult task to predict stock prices because, in every second, the market price is fluctuating. Most of the investors have an interest in doing research on prediction of price financial products such as gold, mutual fund, crude oil, currency exchange, and minerals using varieties of machine learning and data mining techniques. On the basis of different parameters of data like opening price, closing price, date, high and low, stock market price has been predicted. Basically for Indian markets, the two indices such as Sensex for BSE (Bombay Stock Exchange) and nifty for NSE (National Stock Exchange) are the benchmark indices used for forecasting the market price. Day by day huge volumes of people are paying their keen interest in money trading and they want to become profit makers instantly. Here we proposed a novel rough set—Support Vector Machine (R-SVM) approach to predict the Indian stock market data. The R-SVM method is found to be prominent when compared with rough set based algorithms such as decision tree, Naïve Bayes, and artificial neural network in terms of prediction accuracy and complexity.KeywordsBSENSERough setSVMMin-maxPrincipal component analysis