Pemodelan Regresi Spasial untuk Menentukan Faktor-Faktor yang Berpengaruh terhadap Tingkat Kriminalitas di Provinsi Bali dan Jawa Timur

I Putu Gde Inov Bagus Prasetya (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Halu Oleo, Kendari)
Baharuddin Baharuddin (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Halu Oleo, Kendari)
Gusti Ngurah Adhi Wibawa (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Halu Oleo, Kendari)

jurnal syntax admiration
Vol. 5 No. 6 (2024): Jurnal Syntax Admiration

The aim of this research is to design a spatial regression model with an inverse distance weighting matrix formed from the crime rates in the provinces of Bali and East Java, based on influencing factors, and to explore the interconnection between crime rates in one district/city with others. This study utilizes crime rate data obtained from the 2023 publications of the Statistics Indonesia (BPS) of Bali and East Java, covering 47 districts/cities. Spatial regression with inverse distance matrix is employed to analyze the factors influencing crime rates in both provinces, including spatial autocorrelation. The research findings indicate interconnection among districts/cities in the two provinces. The Spatial Error Model (SEM) shows that factors such as average length of schooling, population density, gender ratio, and GDP per capita significantly influence crime rates. This model has an AIC value of 525.06 and a pseudo-R2 value of 64.10%.