Pemodelan Dan Pemetaan Jumlah Property Crime Di Provinsi Jawa Timur Dengan Geographically Weighted Negative Binomial Regression Dan Flexibly Shaped Spatial Scan Statistic

Tambunan, Erika Rindang Kasih (2025) Pemodelan Dan Pemetaan Jumlah Property Crime Di Provinsi Jawa Timur Dengan Geographically Weighted Negative Binomial Regression Dan Flexibly Shaped Spatial Scan Statistic. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043211030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (4MB) | Request a copy

Abstract

Kriminalitas menjadi salah satu ancaman signifikan bagi keamanan masyarakat, dengan risiko yang bervariasi di setiap wilayah. Untuk menganalisis variabel prediktor yang memengaruhi jumlah kasus property crime di Provinsi Jawa Timur, diperlukan pemodelan yang tepat guna mendukung upaya penurunan kasus secara efektif. Perbedaan kondisi ekonomi, pendidikan, dan sosial antar daerah dapat menyebabkan efek spasial. Salah satu metode yang dapat digunakan adalah Geographically Weighted Negative Binomial Regression (GWNBR). Selain itu, pemetaan kasus kejahatan properti dilakukan menggunakan Flexibly Shaped Spatial Scan Statistic untuk mengidentifikasi wilayah dengan tingkat kejadian tinggi. Hasil penelitian menunjukkan bahwa jumlah kasus property crime paling banyak ditemukan di Kota Surabaya dan yang paling rendah di Kabupaten Pacitan. Pemodelan jumlah kasus property crime dengan metode GWNBR menunjukkan bahwa terdapat dua kelompok kabupaten/kota berdasarkan variabel signifikan yang memengaruhi jumlah kasus kejahatan properti. Pemodelan terbaik dengan nilai AIC terkecil sebesar 425,25 yaitu GWNBR dengan pembobot Adaptive Bisquare. Secara keseluruhan, terdapat empat variabel signifikan yang memengaruhi jumlah kasus property crime, yaitu indeks keparahan kemiskinan (X1) yang menunjukkan semakin tinggi keparahan kemiskinan, semakin tinggi risiko kejahatan; ketahanan pangan (X2) yang rendah mendorong peningkatan kasus; persentase rumah tangga tidak menerima bantuan subsidi pemerintah (X4) yang berhubungan dengan keterbatasan dukungan ekonomi; dan laju pertumbuhan PDRB (X6) yang rendah berdampak pada terbatasnya lapangan kerja. Deteksi hotspot property crime mengidentifikasi lima kluster, dengan Kota Mojokerto sebagai daerah paling rawan.
====================================================================================================================================
Crime is one of the significant threats to public safety, with risks varying in each region. To analyze the predictor variables that affect the number of property crime cases in East Java Province, appropriate modeling is needed to support efforts to reduce cases effectively. Differences in economic, educational, and social conditions between regions can cause spatial effects. One method that can be used is Geographically Weighted Negative Binomial Regression (GWNBR). In addition, the mapping of property crime cases was carried out using Flexibly Shaped Spatial Scan Statistics to identify areas with high incidence rates. The results of the study show that the number of property crime cases is most found in the city of Surabaya and the lowest in Pacitan Regency. Modeling the number of property crime cases using the GWNBR method shows that there are two groups of districts/cities based on significant variables that affect the number of property crime cases. The best modelling with the smallest AIC value of 425.25 is GWNBR with an Adaptive Bisquare weight. Overall, there are four significant variables that affect the number of property crime cases, namely the poverty severity index (X1) which shows that the higher the severity of poverty, the higher the risk of crime; low food security (X2) drives an increase in cases; the percentage of households that do not receive government subsidy assistance (X4) related to limited economic support; and the low GDP growth rate (X6) has an impact on limited employment. Property crime hotspot detection identified five clusters, with Mojokerto City as the most vulnerable area.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Flexibly Shaped Spatial Scan Statistic, GWNBR, Jawa Timur, Jumlah Property Crime, Regresi Binomial Negatif, East Java, Flexibly Shaped Spatial Scan Statistic, GWNBR, Negative Binomial Regression, Number of Property Crime
Subjects:	H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics H Social Sciences > HA Statistics > HA30.6 Spatial analysis H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. Q Science > QA Mathematics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics) Q Science > QA Mathematics > QA401 Mathematical models.
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Erika Rindang Kasih
Date Deposited:	05 Feb 2025 13:41
Last Modified:	05 Feb 2025 13:41
URI:	http://repository.its.ac.id/id/eprint/118309

Actions (login required)

View Item