Determination of the Optimal Model on Money Supply Data in Indonesia Using Backpropagation Neural Networks

Abstract

Inflation is a very important economic indicator, the rate of growth is always kept low and stable so as not to cause macroeconomic diseases which will later have an impact on economic instability. The money supply is an indicator that influences the inflation rate, so that the government can control the inflation rate by stabilizing the amount of money in circulation through the monetary policy of Bank Indonesia. The purpose of this research is to predict the amount of money using a backpropagation neural network model, to know the level of accuracy of the model and to predict the money supply in the future. The optimal model of the money supply obtained from the application of the backpropagation neural network is ANN BP(12,6,1). The modeling accuracy obtained from the optimal model of ANN BP(12,6,1) is 92.47% or with a MAPE value of 7.53%. The BP(12,6,1) model is categorized as a model that has very good forecasting capabilities, so it is appropriate to be used as a reference model for forecasting data on the amount circulating in Indonesia in the future.