Time series forecasting remains a challenging task owing to its nonlinear, complex and chaotic behavior. Autoregressive integrated moving average (ARIMA) models are most frequently used since long time in forecasting. Artificial neural networks (ANN) is considered a good alternative to traditional ARIMA model in time series forecasting and often regard superior than ARIMA in forecasting performance. In recent literature Support vector machines (SVM) is becoming famous for solving nonlinear regression problems and time series forecasting. In this study, a hybrid methodology is used which combines the linear ARIMA with nonlinear models of ANN and SVM in order to improve the forecasting performance of Pakistan’s macroeconomic variables such as inflation, exchange rate and stock return. The forecasting performance of all models i.e., ARIMA, ANN, SVM, ARIMA-ANN and ARIMA-SVM are compared on the basis of RMSE and MAE. The results indicate that the best forecasting model to achieve high forecast accuracy is the hybrid ARIMA-SVM. Supervisor:-Dr. Hafsa Hina Co-Supervisor:- Dr. Amena Urooj