This analysis compares model performance of longitudinal beta regression using GLMM and GEE, highlighting implications for economic inequality assessment.
Key Points
GLMM outperforms GEE in longitudinal beta regression, providing a better fit for bounded outcome data.
The study reports consistently lower RMSE and higher pseudo R-squared for GLMM in modeling economic inequality.
Assessment used panel data from 10 provinces, focusing on socioeconomic indicators between 2018 and 2024.
Findings suggest that modeling techniques, rather than data heterogeneity, influence performance outcomes.