This article investigates the issue of real versus nominal adjustment mechanism in the Malaysian demand for money function. The test for stability and ex-post forecasting ability of the model are also conducted. The results suggest that for Ml, both specifications are the 'true' model, and for M2 and M3, the nominal specification is the 'true' model. The stability test indicates that only the real specification for M2 and M3 are stable, and the results of the ex-post forecast suggest that the nominal specification has a better forecasting ability compared to the real specification of the Malaysian money demand function.