Analysis of Indonesia’s Gross Regional Domestic Product using a Spatially Filtered Unconditional Quantile Regression Approach

Sri Amaliya (Department of Statistics, IPB University)
Anik Djuraidah (Department of Statistics, IPB University)
Budi Susetyo (Department of Statistics, IPB University)

JTAM (Jurnal Teori dan Aplikasi Matematika)
Vol 10, No 3 (2026): July

Analyzing Indonesia’s Gross Regional Domestic Product (GRDP) is crucial for understanding regional economic disparities characterized by heterogeneity and spatial dependence. However, previous studies using mean regression or standard Unconditional Quantile Regression (UQR) often ignore spatial dependence, potentially biasing distributional estimates. Substantively, this study aims to examine how socioeconomic factors influence regional economic performance across different levels of GRDP in Indonesia. To address the methodological gap, this study applies Spatially Filtered Unconditional Quantile Regression (SF-UQR), which captures heterogeneous effects across the GRDP distribution while accounting for spatial dependence. Using cross-sectional data of Indonesian districts and cities from Statistics Indonesia (BPS) in 2023, GRDP is specified as the response variable, with five explanatory variables: human development index, minimum wage, number of workers, original local government revenue, and poverty rate. The analysis compares UQR and SF-UQR across selected quantiles. The results reveal substantial heterogeneity. Human development index and original local government revenue consistently show positive effects, poverty rate negatively affects lower quantiles, minimum wage exhibits a shifting pattern, and number of workers is significant mainly at middle and upper quantiles. SF-UQR outperforms standard UQR, achieving an adjusted R² of 0,67 compared to 0,52 under UQR. Methodologically, this study highlights the relevance of incorporating spatial filtering into UQR when analyzing regional economic data characterized by spatial dependence, providing an alternative distributional perspective on regional economic dynamics. From a policy perspective, the findings indicate that development strategies should consider both distributional heterogeneity and spatial dependence. Overall, the results highlight the necessity of spatially informed and distribution-sensitive policy design to reduce regional economic disparities in Indonesia.