PROJECTION OF INFRASTRUCTURE BUDGET REQUIREMENTS USING TIME SERIES ANALYSIS

Abstract

In implementing the mandate of the 1945 Constitution, the Government uses the State Revenue and Expenditure Budget (APBN) instrument to realize the state's goals, namely advancing the general welfare of the entire Indonesian nation. The structure of state spending in the APBN consists of central government spending and transfer spending to regions. Infrastructure spending is state spending that has a large portion in the APBN because it is direct spending related to accelerating the development of public service facilities and the economy. In addition, the allocation of infrastructure spending since 2014 has continued to increase, thereby reducing fiscal space and the portion of spending for other priority activities. Therefore, it is necessary to accurately calculate infrastructure spending needs for the following year so that it can be allocated effectively and efficiently. The data used in projecting infrastructure spending for 2025 is annual data for 11 years. The data is divided into three parts according to the conditions of each data. The results of the stationary test show that the data used is not stationary in terms of variance so that not all time series analysis models can be used. Based on the results of several time series analysis models, the single exponential smoothing model on the second data has the smallest MAPE value, namely 9.8. From this model, a projection for infrastructure spending in 2025 is IDR 425.557 trillion.