Pemodelan Spasial Tingkat Pengangguran Terbuka Kabupaten/Kota di Pulau Sumatera Menggunakan Pendekatan Geographically Weighted Panel Regression

Abstract

This study aims to analyze the factors affecting the Open Unemployment Rate (OUR) at the regency/city level in Sumatra Island during the 2022–2024 period by considering spatial and temporal variations. The study employs panel data obtained from Statistics Indonesia, with the Open Unemployment Rate as the dependent variable and the Human Development Index, Mean Years of Schooling, Labor Force Participation Rate, and Gross Regional Domestic Product as independent variables. Panel data regression and Geographically Weighted Panel Regression (GWPR) are applied as analytical methods. The best panel regression model is selected using the Chow and Hausman tests, which indicate that the Fixed Effect Model is the most appropriate. The Breusch–Pagan test confirms the presence of spatial heterogeneity, justifying the use of the GWPR approach. The GWPR model with an adaptive bisquare kernel weighting function and optimal bandwidth successfully captures local variations in the relationship between explanatory variables and unemployment. The results reveal that Mean Years of Schooling and Gross Regional Domestic Product have a positive effect on the Open Unemployment Rate, while the Labor Force Participation Rate has a negative effect, with varying magnitudes across regions. The GWPR model outperforms the global panel regression, achieving a coefficient of determination of 96.8%. These findings highlight the importance of incorporating spatial approaches in formulating region-specific employment policies in Sumatra Island.