Regresi Nonparametrik Spline Truncated untuk Memodelkan Tingkat Kriminalitas di Sumatera Utara

Zahrah, Aisyah Maulidya Cahya Az (2025) Regresi Nonparametrik Spline Truncated untuk Memodelkan Tingkat Kriminalitas di Sumatera Utara. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003211067-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Kriminalitas merupakan isu utama yang dihadapi masyarakat, baik di negara maju maupun berkembang. Berdasarkan data Statistik Kriminal 2024, jumlah kejahatan di Indonesia mengalami peningkatan, dengan Provinsi Sumatera Utara menduduki peringkat ketiga tertinggi secara nasional. Meskipun persentase penyelesaian kasus meningkat pada tahun 2023, risiko penduduk terkena kejahatan tetap tinggi. Selaras dengan tujuan Sustainable Development Goals (SDGs) 2030, khususnya dalam menciptakan masyarakat yang aman dan sejahtera, penelitian ini bertujuan untuk menganalisis faktor-faktor yang diduga mempengaruhi tingkat kriminalitas di Sumatera Utara menggunakan metode regresi nonparametrik spline truncated. Metode ini dipilih karena pola hubungan antara tingkat kriminalitas dan faktor-faktor yang diduga mempengaruhinya tidak mengikuti pola data tertentu. Hasil penelitian menunjukkan bahwa model terbaik yang diperoleh adalah model dengan kombinasi knot (3,3,2,3) yang ditentukan berdasarkan nilai Generalized Cross Validation (GCV) terkecil. Dengan menggunakan empat variabel prediktor yaitu kepadatan penduduk, persentase penduduk miskin, rata-rata lama sekolah, dan produk domestik regional bruto per kapita atas dasar harga konstan diketahui bahwa seluruh variabel tersebut berpengaruh signifikan terhadap tingkat kriminalitas di Sumatera Utara. Nilai koefisien determinasi yang dihasilkan dari model adalah sebesar 92,61%.
===========================================================================================================================================
Crime is a major issue faced by society, both in developed and developing countries. Based on the 2024 Crime Statistics data, the number of crimes in Indonesia has increased, with North Sumatra Province ranking third highest nationally. Although the percentage of case resolution increased in 2023, the risk of residents being victims to crime remains high. In line with the objectives of the 2030 Sustainable Development Goals (SDGs), especially in creating a safe and prosperous society, this study aims to analyze the factors that are estimated to influence the crime rate in North Sumatra using the Nonparametric Spline Truncated Regression method. This method was chosen because the pattern of the relationship between the crime rate and the factors suspected of influencing it does not follow a particular data pattern. The results of the study showed that the best model obtained was a model with a combination of knots (3,3,2,3) which was determined based on the smallest Generalized Cross Validation (GCV) value. By using four predictor variables, namely population density, percentage of poor people, average length of schooling, and gross regional domestic product per capita at constant prices, it is known that all of these variables have a significant effect on the crime rate in North Sumatra. The coefficient of determination value resulting from the model is 92,61%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Generalized Cross Validation, Regresi Nonparametrik Spline Truncated, Sumatera Utara, Tingkat Kriminalitas, Titik Knot, Crime Rate, Generalized Cross Validation, Knot Points, North Sumatra, Truncated Spline Nonparametric Regression
Subjects:	H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Aisyah Maulidya Cahya Az Zahrah
Date Deposited:	01 Aug 2025 04:15
Last Modified:	01 Aug 2025 04:15
URI:	http://repository.its.ac.id/id/eprint/125528

Actions (login required)

View Item