Metode Autoregressive Integrated Moving Average (Arima) Untuk Meramalkan Jumlah Uang Beredar (M2) Di Indonesia

Abstract

Main focus of this research is how business could predict amount of money supply in the market. Bussiness interests are depend on inflation rate in the macro economy. Money supply plays important role to determine of inflation rate in the region or country. When money was over supply than availability of good and services, demand of good and services would stress the market price (inflation). Otherwise, when money supply was less than availability of good and services, demand of good and services would be decreasing (deflation). Bank Indonesia as a monetary authority had the monetary policy to stabilize amount of money supply.  

ARIMA or Box Jenkins approach has been chosen as research method in this research. Estimation of ARIMA model is just based on secondary data of single variable which is analyzed. The data of money supply was collected based on publication of Bank Indonesia. ARIMA model is a method to form regression estimation model based on serial of data itself.  ARIMA method had standard procedures to develop the model estimation. Four standard procedures are identification, estimation, diagnostic checking, and forecasting. Final result of this model is regression estimation which consist of money supply as dependent variable and autoregressive of data itself and error term (moving averages) as independent variables.

Based on the analyzes shows that best model is DLM2 as dependent variables (money supply after second deferenciated) and independent variables that consist of autoregressive (AR) on fifth order or AR (5) dan moving average (MA) on fifth order or MA (5). Therefore, estimation of money supply model is ARIMA (5,1,5) and this model is sufficient to forecast money supply in the future because it has low rate of RMSE.

Keywords: money supply, inflation, ARIMA.