Forecasting the Amount of Cash in Circulation Based on a Regression Model with Lagged Variables

Abstract

Despite the rather positive dynamics in developing the non-cash payment infrastructure, cash in Russia has been and remains the widespread instrument of payment. The purpose of this paper is the development of a model for forecasting the amount of cash in circulation in the country, namely the value of the monetary aggregate M0, by the example of the Russian Federation. At the same time, the objectives were set so that the resulting model should be suitable for a sufficiently accurate rapid estimation of М0 value, should be easy to use, and should not be overloaded with many variables. The author succeeded in achieving these purposes in the research using a formal approach based on a model with lagged variables, autoregression and time series analysis, also using numerical methods. The lagged variables used were the inflation rate and the value of the monetary aggregate for a previous period. The high quality, accuracy and forecasting power of the model are substantiated. The average approximation error of the model did not exceed 5%.Testing of the model using statistical data of the current year showed high accuracy of forecasting. Statistical data of official sources of the Russian Federation were used for the model development