Model Data Panel Untuk Menganalisis Kontribusi Sektor Pariwisata Terhadap Pendapatan Asli Daerah (PAD) Di Kabupaten/Kota Provinsi Sumatera Utara Tahun 2020-2023

Abstract

Regional fiscal independence is reflected in local own-source revenue (PAD). The tourism sector plays an important role in the stability of local revenue, but the spatial and temporal heterogeneity between districts/cities in North Sumatra makes it difficult to analyze the contribution of this sector using ordinary linear regression. Therefore, a panel data statistical approach is needed to comprehensively identify the dynamics of this relationship. This study aims to examine the influence of the tourism sector on PAD in the districts/cities of North Sumatra Province for the period 2020-2023. This study also evaluates the effectiveness of panel data regression models (Common Effect, Fixed Effect, and Random Effect) in capturing this heterogeneity. This descriptive quantitative method research applies panel data regression using secondary data from the North Sumatra Central Bureau of Statistics (BPS). The dependent variable is PAD, while the independent variables include the number of hotels and accommodations room occupancy rate and average length of guest stay. Optimal model selection was conducted through the Chow test, Hausman test, and Lagrange Multiplier test. Furthermore, classical asumption (normality, multicollinearity, heteroscedasticity) and hypothesis testing (partial and simultaneous). The Fixed Effect Model (FEM) was selected as the best model based on the Chow Test and Hausman Test. FEM was able to explain variations in PAD with an R-Squared value of 0.9743 or 97.43%. Partially, only the variable number of hotels and accommodation (X1), had a significant negative effect on PAD (coefficient-5760487; p-value 0.0000). The variables of room occupancy rate (X2) and average length of guest stay (X3) showed no significant effect. Classical assumption testing shows that the data is normally distributed, no heteroscedasticity, and no multicollinearity. This study shows that the tourism sector contributes significantly to PAD, but its effectiveness is highly dependent on the quality and distribution of tourism facilities. An increase in the number of accommodations without being supported by the quality of services does not necessarily encourage an optimal increase in PAD.