Modelling currency in circulation in India

In the literature, currency in circulation is typically estimated either by specifying a standard currency demand equation based on the theory of transaction or portfolio demand for money or by a univariate time series model. The paper argues that while these approaches work well for low frequency data, their scopes become limited for high frequency series. Using the weekly data of currency in circulation from the year 1992-1993 to 1999-2000, the paper proposes an alternative approach of modelling the growth of currency in circulation by incorporating the 'day of the month' effect. The estimated equation behaves very well in and out of sample and the prediction performance achieved from the model has been good.