<![CDATA[This study aims to analyze the effect of zakat distribution on the Human Development Index (HDI) in Indonesia by comparing two modeling approaches: the Mixed Effect Model (LMM) and Extreme Gradient Boosting (XGBoost). The data used are panel data from 34 provinces during the 2021–2024 period, which includes sectoral zakat distribution and HDI variables. Initial exploration results indicate spatial variation and a non-linear correlation between zakat distribution and HDI. The Mixed Effect model demonstrated more stable predictive performance with an RMSE of 1.633 and an R² of 0.762 on the test data. While XGBoost, despite its high accuracy on the training data, exhibited overfitting with a decrease in accuracy on the test data. These findings indicate that the LMM approach is superior in capturing spatial and temporal variations between regions and provides more applicable interpretations in the context of public policy. This study contributes to the use of statistical approaches and machine learning to evaluate the effectiveness of zakat in supporting sustainable human development in Indonesia.]]>
Copyrights © 2025
