Forecasting is an important task in stock markets, and it has held the attention of academics and practitioners over the last two decades. The extensive research reflects the importance of volatility in investment, security valuation, risk management, and monetary policy decision making. Predicting daily behaviour of stock market is a challenge for investors and corporate stockholders and it can help them to invest with more confident by taking risks and fluctuations into consideration. In this paper, we show the various forecast procedure like Time series Prediction Methods (Random Walk, Regression Method and ARIMA Models) and Machine Learning Method (Artificial Neural Network) to predict the future values. The performance of the models is evaluated by calculating various statistical measures like Mean Error (ME), Mean Absolute Error (MAE), Average Absolute Error (AAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Percentage Square Error (MPSE) and statistical indicators like Autocorrelation, Correlation Coefficient, Mean Absolute Deviation, Squared Correlation, Standard Deviation and compare their values and to compare the results and trends of actual and predicted values of above mentioned indices by Diebold Mariano test, Akaike’s minimum final prediction error (FPE), Theil Inequality coefficient, Goodness of fit, t-test, and Mariano’s test for significant difference.