In the business sector, it is always a difficult task to predict the daily prices of stock market due to its nonlinear behavior in stock market prices. Hence, there are numerous study has been conducted in respect of the prediction of the direction of stock market movement. Several studies provide solid corroboration that models using traditional regression techniques face irresistible problems due to model ambivalence. So here there is a need of discriminating model used to minimize the high risk and maximize returns. Computing techniques are rational method for forecasting stock prices. This study presents a computational approach for predicting the stock market indexes of six different countries. It is a review of artificial neural network (ANN) and support vector machine (SVM) pertain and achieve to predict the stock market prices. Daily stock price data from 2000 to 2018 has been used for prediction. Six stock market indexes of Asian region have been selected to predict the daily stock prices by using 30 variables. The proposed methods consists of three steps in first step it uses technical analysis based on historical data to calculate important indicators. In second step it identified the most important indicators which actually represents the data set and then finally use ANN and SVM to predict the stock market prices by using important variables. For better performance, the study have performed comparative analysis to find out better model in each country. The results concluded that SVM and ANN has ability to predict the stock market and the main finding of the study is that by using the most important feature extraction the prediction accuracy has been increased and as a result investors can earn huge profit by using these techniques. Supervisor: Dr. Faheem Aslam Co-Supervisor: Dr. Saud Ahmed Khan