Forecasting Value-Added Tax (VAT) revenue using Autoregressive Integrated Moving Average (ARIMA) Box-Jenkins method

Abstract

We propose a method to forecast Value-Added Tax (VAT) revenue for Indonesia government using Autoregressive Integrated Moving Average (ARIMA) Box-Jenkins method. We experimented the ARIMA Box-Jenkins method using time-series analysis of VAT revenue data of two Tax Offices of Directorate General of Taxes (DGT) from the last five years. The result shows that it resembles the real VAT revenue more closely than when compared to the actual VAT target by Indonesia government. We then argue that this result may be used as a fail-safe tax revenue target, that can work as a tool to better measure DGT performance.