PEMODELAN GEOGRAPHICALLY WEIGHTED REGRESSION PRINCIPAL COMPONENT ANALYSIS (GWR-PCA) DENGAN FUNGSI PEMBOBOT FIXED GAUSSIAN KERNELNIM.: 22106020073Redella Reffa HerdiantiauthorGeographically Weighted Regression Principal Component Analysis (GWRPCA) is a combination of Principal Component Analysis (PCA) and Geographically Weighted Regression (GWR). This study aims to model the factors affecting the distribution percentage of Gross Regional Domestic Product (GRDP) across regencies/cities in South Sulawesi Province in 2024 using GWRPCA with a fixed Gaussian kernel weighting function. PCA was employed to address multicollinearity among independent variabels, while spatial heterogeneity was examined using the Breusch–Pagan test. The optimum bandwidth was determined using the Cross Validation (CV) method, and parameter estimation was carried out using the Weighted Least Squares (WLS) method. The PCA results produced two principal components, namely PC1 and PC2, which explained 66.27% of the total data variance. The Breusch–Pagan test indicated the presence of spatial heterogeneity (p-value = 0.003), supporting the application of the GWRPCA model. The GWRPCA model yielded an R² value of 0.8170, an AIC value of 120.7012, and an RMSE value of 2.8015, outperforming the RPCA model. The results showed that PC1 and PC2 had a significant effect on the distribution percentage of GRDP across all regencies/cities in South Sulawesi Province.510 Mathematics (Matematika)2026-06-22UIN SUNAN KALIJAGA YOGYAKARTA;FAKULTAS SAINS DAN TEKNOLOGIThesis