This study analyzed the link between the PKR-US dollar currency exchange rate determination and the monetary policy defined by Taylor rule. This study also incorporate the non-linearity linked to the exchange rate behavior in the Taylor rule (TR) exchange rate model by employing the Smooth Transition Auto Regression technique (STAR) over the period of 1982Q1-2019Q4. The forecasting ability of the non-linear TR model is assessed relative to the benchmark random walk model by different forecast evaluation techniques which include Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Theil’s U. The linearity test results show the non-linear association between the nominal exchange rate and the Taylor rule variables. The 1st order logistic STAR model is found to be the most appropriate for modeling the PKR-US dollar exchange rate behavior. The estimation results of the forecast accuracy tests i.e RMSE, MAE and Theil’s U coefficient show that the non-linear TR exchange rate model have high forecasting power as compared to the benchmark random walk at all forecast horizon. Supervisor:- Dr. Nasir Iqbal