The knowledge of money demand function of an economy is indispensible for a successful monetary policy and its crucial to know the appropriate monetary aggregate that should be used in money demand model. The relative performance and efficacy of simple sum monetary aggregates are compared with that of divisia sum monetary aggregates in the context of money demand function for Pakistan by considering the impact of financial liberalization as well. Time series data is collected from 1981 to 2014 for the case of Pakistan. Simple sum and Divisia sum Monetary aggregates were created using the methodology presented by Barnett (1980). ARDL approach to co-integration is applied to estimate the long run and short run impact of US Treasury bill rate, domestic saving rate, nominal effective exchange rate and real income and money demand. The Toda-Yamamoto approach to Granger causality has also been applied. Monetary aggregate M2 of both simple sum and divisia sum approach are found to be statistically insignificant in the money demand model. Although, both simple sum and divisia sum monetary aggregates Mo and M1 found to be significant however monetary aggregate M1 for both simple sum and divisia sum approach is found to be marginally better than their Mo counterpart. Yamamoto Granger Causality results also verified the causality from the US Treasury bill rate, nominal effective exchange rate and real income towards the simple sum and divisia sum Mo and M1. The result of the study suggest that policy makers may also construct, report and use divisia sum money measures for Pakistan along with simple sum counterparts owing to the fact that they contain relatively more information and performs marginally better than simple sum aggregates. Supervisor: Dr. Attiya Yasmin Javid