<![CDATA[The total number of crimes in East Java Province reached 60,102 cases in 2024, indicating spatial and temporal variation in crime patterns. These data characteristics are heterogeneous and contain several extreme values (outliers) that may affect the stability of standard regression model estimates. This study applies the Robust Mixed Geographically and Temporally Weighted Regression (RMGTWR) model to examine spatiotemporal crime patterns in East Java during the 2020–2024 period and to evaluate the model’s ability to handle outliers in heterogeneous data. The evaluation results show that the RMGTWR model produced an of 0.84 and an MSE of 1912.247. This robust approach provides more stable parameter estimates and reduces the influence of the 14 identified outlier observations compared with the non-robust model. The findings indicate that the Human Development Index (HDI) and the Gini Ratio have significant global effects, while the Police Ratio variable shows varying local effects. These results confirm that the RMGTWR approach accommodates data heterogeneity while mitigating distortions caused by outliers in crime analysis in East Java.]]>
Copyrights © 2026
