Puspa Nur Afifah
UIN SULTAN MAULANA HASANUDDIN BANTEN

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Evaluasi Model Regresi GWR dan NBR untuk Analisis Kriminalitas di Banten: Tantangan Sampel Terbatas Puspa Nur Afifah; Sekar Ayu Permata; Wahyu Arjuna Saputra
Teorema: Teori dan Riset Matematika Vol 10, No 2 (2025): September
Publisher : Universitas Galuh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25157/teorema.v10i2.19317

Abstract

Penelitian ini mengevaluasi tiga model regresi dalam menganalisis tingkat kriminalitas di Provinsi Banten, yaitu Regresi Poisson, Regresi Geographically Weighted Regression (GWR), dan Regresi Binomial Negatif (NBR). Data mencakup delapan wilayah administratif dengan variabel penjelas seperti kepadatan penduduk, rasio penduduk terhadap polisi, tingkat pengangguran terbuka, angka kemiskinan, rata-rata lama sekolah, dan Indeks Pembangunan Manusia. Hasil menunjukkan bahwa 1) model Poisson mengalami overdispersi berat dan gagal menangkap pengaruh variabel prediktor, 2) model GWR memiliki nilai AIC terendah, namun menghasilkan parameter yang identik di semua wilayah, mengindikasikan overfitting akibat keterbatasan jumlah wilayah, 3) Model NBR memberikan deviasi rendah dan AIC yang kompetitif, menunjukkan kinerja statistik yang lebih stabil pada data terbatas. Studi ini menyoroti pentingnya mempertimbangkan ukuran sampel dalam pemilihan model spasial, serta menyarankan penggunaan NBR sebagai alternatif yang layak untuk data count yang overdispersi dengan jumlah unit kecil.
Investigating Employment Delays: A Multiple Classification Analysis of Job Search Duration among Graduates in Islamic Community Development Puspa Nur Afifah; Nadya Rahmadhani; Syehabu Rifqi
KUBIK Vol 9 No 2 (2024): KUBIK: Jurnal Publikasi Ilmiah Matematika
Publisher : Department of Mathematics, Faculty of Science and Technology, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/kubik.v9i2.36256

Abstract

Indonesia is facing big challenges: an overabundance of graduates and limited job prospects. In February 2022, the central statistics agency provided a clear picture, more than 13 million people have a bachelor's degree in Indonesia and the number of unemployed is 8.4 million people. Banten Province is ranked first in the unemployment rate in Indonesia, namely 8.53%. The aim of this research is to determine the factors that influence the length of job search for graduates of the Islamic Community Development Program at UIN Sultan Maulana Hasanuddin Banten. The analytical method used is Multiple Classification Analysis (MCA), this research investigates the influence of variables, such as gender, length of study, GPA, organizational experience, academic achievement, work experience, employment sector, and foreign language proficiency. The research results show that there is a significant relationship between length of study, GPA, work experience, and length of job search. The ANOVA test shows that these three variables have a low significance level (