PREDICTION OF THE VALUE OF THE REGIONAL REVENUE AND EXPENDITURE BUDGET (APBD) OF NORTH SUMATERA PROVINCE USING THE POLYNOMIAL REGRESSION METHOD
Keywords:
Keywords: APBD, Polynomial Regression, Prediction, MAPE, RMSE, North SumatraAbstract
The Regional Revenue and Expenditure Budget (APBD) of North Sumatra Province experiences fluctuations in value from year to year, so a method is needed that is able to predict the value of the APBD in the future to assist regional budget planning. This study aims to apply the Polynomial Regression method in predicting the value of each APBD account of North Sumatra Province and evaluate its accuracy level using the Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) metrics. The data used are historical data of the APBD revenue and expenditure section for 2005–2025 which consists of seven accounts, namely Regional Original Revenue, Transfer Revenue, Other Legitimate Revenue, Operational Expenditure, Capital Expenditure, Unexpected Expenditure, and Transfer Expenditure. The data is divided into training data (2005–2021) and test data (2022–2025). The Polynomial Regression method is tested at degrees 1, 2, and 3, with the best model selected based on the smallest MAPE value. The results of the study show that the best model for each account varies, with Local Original Revenue obtaining a grade 2 model (MAPE 8.23%, RMSE 899.38 billion Rp), Transfer Revenue grade 1 (MAPE 53.65%), Other Legitimate Income grade 1 (MAPE 1,609.59%), Operational Expenditure grade 3 (MAPE 45.91%), Capital Expenditure grade 1 (MAPE 11.79%), Unexpected Expenditure grade 3 (MAPE 33.76%), and Transfer Expenditure grade 1 (MAPE 8.96%). The 2026–2030 Regional Budget (APBD) projections show an upward trend in most accounts, particularly Regional Original Revenue (PAD), which is projected to increase from Rp 7,890.03 billion in 2026 to Rp 8,820.80 billion in 2030. The Polynomial Regression model is quite reliable for accounts with stable trends (MAPE <20%), but is less recommended for accounts with high fluctuations.